System

Abstract

To provide a system.SOLUTION: A system includes: proposal means for generating one or a plurality of mailing destination candidates of a delivery on the basis of a sender's name, the sender's address, the type of the delivery from the sender, the size of the delivery and the format of the delivery upon entry of a specific prompt statement into a generative AI model, and for proposing one or the plurality of mailing destination candidates that is generated to a user who is an assortment operator; an emotion engine for recognizing the user's emotion when the mailing destination candidates of the delivery are proposed; and adjustment means for adjusting the number of the mailing destination candidates on the basis of the emotion.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including a description and related instruction sentence regarding the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] When sorting mail and parcels with unknown addresses, there is an issue that the workload is heavy due to the need to check the sorting list each time and the work experience of the person in charge. [Means for solving the problem]

[0005] The system disclosed herein includes a suggestion means for generating one or more candidate destinations for the delivery item based on the name of the sender, the address of the sender, the type of delivery item from the sender, the size of the delivery item, and the format of the delivery item by inputting a specific prompt sentence into a generative AI model, and suggesting the generated one or more candidate destinations to a user who is a sorter; an emotion engine for recognizing the emotions of the user who has been suggested the candidate destinations for the delivery item; and an adjustment means for adjusting the number of candidate destinations based on the emotions. [Brief description of the drawings]

[0006] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Diagram 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Diagram 5] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 2 is a sequence diagram showing a process flow of the data processing system in the first embodiment of the first form example. [Figure 12] FIG. 11 is a sequence diagram showing a process flow of the data processing system in the application example 1 of the first embodiment. [Figure 13] FIG. 11 is a sequence diagram showing a process flow of the data processing system in the second embodiment of the second embodiment. [Figure 14] FIG. 13 is a sequence diagram showing a process flow of the data processing system in application example 2 of embodiment 2. [Figure 15]FIG. 13 is a sequence diagram showing a process flow of a data processing system in a third embodiment of the present invention. [Figure 16] FIG. 13 is a sequence diagram showing a process flow of the data processing system in the third application example of the third embodiment. [Figure 17] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when combined with an emotion engine. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0008] First, the terms used in the following description will be explained.

[0009] In the following embodiments, a signed processor (hereinafter simply referred to as a "processor") may be one arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be one type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] In the following embodiments, a signed RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by the processor.

[0011] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0012] In the following embodiments, a communication I / F (Interface) with a code is an interface including a communication processor and an antenna. The communication I / F controls communication between multiple computers. An example of a communication standard applied to the communication I / F is a wireless communication standard including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "and / or."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0017] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a wide area network (WAN) and / or a local area network (LAN).

[0018] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0019] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (e.g., a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0020] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (e.g., voice and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs voice according to instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a Complementary Metal-Oxide-Semiconductor (CMOS) image sensor or a Charge Coupled Device (CCD) image sensor.

[0021] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54.

[0022] FIG. 2 shows an example of main functions of the data processing device 12 and the smart device 14.

[0023] As shown in Fig. 2, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32. The specific process program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific process program 56 from the storage 32, and executes the read specific process program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific process program 56 executed on the RAM 30.

[0024] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores a reception output program 60. The reception output program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads out the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] The system of the present invention is for sorting mail and parcels with unknown addresses. Specifically, it is provided with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. For example, if the sender is "Company A" and the delivery is "Parts Type B," it is suggested to post the delivery to "Department C" based on past data. In addition, if the size of the delivery falls within a specific range, it is possible to suggest possible destinations according to various conditions, such as suggesting posting to a specific location.

[0029] "Example 2"

[0030] The system of the present invention is for sorting mail and parcels with unknown addresses. Specifically, it is provided with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. For example, if the sender is "Company A" and the delivery is "Parts Type B," it is suggested to post the delivery to "Department C" based on past data. In addition, if the size of the delivery falls within a specific range, it is possible to suggest possible destinations according to various conditions, such as suggesting posting to a specific location.

[0031] The process flow of each embodiment will be described below.

[0032] "Example 1"

[0033] Step 1: The system receives information about the bounced mail or parcel, including the sender, type, size, and format of the parcel.

[0034] Step 2: Next, the system proposes possible delivery destinations based on the received information. Specifically, the system generates one or more possible delivery destinations for the delivery item based on the sender and the type, size, and format of the delivery item, and then proposes the generated one or more possible delivery destinations to the user, who is the sorter.

[0035] Step 3: The system then suggests possible delivery destinations based on past delivery destinations and detailed information about the delivery item. For example, if the sender is "Company A" and the delivery item is "Parts Type B," the system will suggest delivery to "Department C" based on past data.

[0036] Step 4: Finally, the system presents the proposed destination candidates to the user, who can select the most suitable destination from the presented destination candidates.

[0037] Example 1

[0038] Next, a description will be given of Example 1 of the embodiment 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal".

[0039] Sorting mail and parcels with unknown addresses requires the staff to check their work experience and sorting list each time, which is inefficient and increases the risk of misdelivery. There is also the problem that it is difficult to quickly find the appropriate delivery destination based on the sender and the type, size, and format of the delivery. To solve these issues, a system is needed that utilizes past data to efficiently and accurately suggest delivery destination candidates.

[0040] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0041] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, a means for generating possible destinations using a generative AI model, a means for transmitting data entered by a user to the server, a means for the server to acquire past data from a database, a means for the server to transmit the proposed results to a terminal, and a means for the user to check the proposed results and select an appropriate destination. This makes it possible to efficiently and accurately sort mail and parcels with unknown addresses.

[0042] The server may include a suggestion means for generating one or more candidate destinations for the delivery item based on the name of the sender, the address of the sender, the type of delivery item from the sender, the size of the delivery item, and the format of the delivery item by inputting a specific prompt sentence into a generative AI model, and suggesting the generated one or more candidate destinations to a user who is a sorter, an emotion engine for recognizing the emotions of the user who has been suggested the candidate destinations for the delivery item, and an adjustment means for adjusting the number of candidate destinations based on the emotions.

[0043] The adjustment means may generate the most likely candidate destination as the candidate destination when the user is feeling stressed, and may generate multiple candidate destinations when the user is relaxed.

[0044] The suggestion means may generate one or more candidate destinations for the piece, including destinations of past pieces by the sender.

[0045] The suggestion means provided in the server may generate one or more candidate destinations for the delivery item based on the name of the sender, the address of the sender, the type of delivery item from the sender, the size of the delivery item, and the format of the delivery item by inputting a specific prompt sentence into a generative AI model, and may further generate one or more candidate destinations for the delivery item including past destinations of deliveries by the sender.

[0046] "Sender" refers to the person or organization sending the mail or package.

[0047] "Delivery" refers to items sent by mail or courier service.

[0048] "Type" is an attribute that indicates the classification of the delivery.

[0049] "Size" is an attribute that indicates the size of the delivery item.

[0050] "Form" is an attribute that indicates the shape of the delivery item and the type of packaging.

[0051] "Possible delivery destinations" refer to locations or departments where deliveries may be dropped off.

[0052] "Previous mailbox location" refers to a location or department where a delivery has been previously mailed.

[0053] "Detailed information" refers to specific data and history about the delivery.

[0054] A "generative AI model" is a model that uses artificial intelligence to analyze data and generate potential recipients.

[0055] A "user" is a person who operates the system to enter postal or courier information.

[0056] A "server" is a computer system that processes and stores data.

[0057] A "terminal" is a computer that includes input and display devices operated by a user.

[0058] A "database" is a system for storing detailed information about past mailing destinations and deliveries.

[0059] "Suggested results" refer to the candidate recipients generated by the generative AI model.

[0060] This invention is a system for efficiently and accurately sorting mail and parcels with unknown destinations. This system is realized using a server, a terminal, a generative AI model, and a database management system (DBMS).

[0061] Hardware and Software Used

[0062] Hardware: Servers, terminals

[0063] Software: Generative AI models, Database Management System (DBMS)

[0064] System Configuration

[0065] 1. User enters postal or courier information:

[0066] The user inputs information about the mail or parcel they received, such as the sender, type, size, and format of the parcel, into the input form on the terminal. For example, the sender is "Company A," the parcel is "Type B parts," the size is "medium," and the format is "box."

[0067] 2. The device sends the input data to the server:

[0068] The terminal sends the data entered by the user to the server, specifically, using an HTTP request to send information such as the sender, type of delivery, size, format, etc.

[0069] 3. The server retrieves historical data from the database:

[0070] The server uses a database management system (DBMS) to retrieve past mailing destinations and delivery details. For example, it executes an SQL query to retrieve past mailing destination data for "Parts of type B" from "Company A."

[0071] 4. The server uses the generative AI model to generate possible recipients:

[0072] The server inputs the acquired past data into the generative AI model to generate candidates for recipients. Specifically, it inputs the following prompt sentence to the generative AI model:

[0073] If the sender is "Company A" and the delivery is "Parts Type B", which department would you suggest to post it to based on past data?

[0074] Based on this prompt, the generative AI model suggests sending the item to "Division C."

[0075] 5. The server sends the proposal results to the device:

[0076] The server sends the candidate mailing destinations generated by the generative AI model to the terminal. Specifically, it uses an HTTP response to send the result proposing mailing to "Department C" to the terminal.

[0077] 6. User reviews the suggestions and selects the appropriate destination:

[0078] The user checks the proposed results displayed on the terminal. For example, if the proposal is to send the item to "Division C," the user can accept the proposal or select another destination. The destination selected by the user is recorded in the system and used as reference data for the future.

[0079] In this way, users can efficiently find the correct drop-off location for undelivered mail or parcels, improving the efficiency of sorting operations and reducing the risk of misdelivery.

[0080] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0081] Step 1:

[0082] A user enters postal or courier information.

[0083] The user inputs information about the mail or parcel they received, such as the sender, type of item, size, and format, into the input form on the terminal. For example, the sender is "Company A," the item is "Type B parts," the size is "medium," and the format is "box." The input data is temporarily stored in the terminal's memory.

[0084] Step 2:

[0085] The terminal transmits the input data to the server.

[0086] The terminal sends the data entered by the user to the server using an HTTP request. Specifically, it sends JSON-formatted data including information such as the sender, type of delivery, size, and format to the server. The server receives and analyzes this data.

[0087] Step 3:

[0088] The server retrieves historical data from the database.

[0089] The server uses a database management system (DBMS) to retrieve past mailing destinations and detailed delivery information. For example, it executes an SQL query to retrieve past mailing destination data for "Part B" from "Company A." The retrieved data is stored in the server's memory.

[0090] Step 4:

[0091] The server uses the generative AI model to generate candidate recipients.

[0092] The server inputs the acquired past data into the generative AI model to generate candidates for recipients. Specifically, it inputs the following prompt sentence to the generative AI model:

[0093] If the sender is "Company A" and the delivery is "Parts Type B", which department would you suggest to post it to based on past data?

[0094] Based on this prompt, the generative AI model suggests sending the mail to "Division C." The generated mail destination candidates are stored in the server's memory.

[0095] Step 5:

[0096] The server transmits the proposal results to the terminal.

[0097] The server sends the candidate mailing destinations generated by the generative AI model to the terminal using an HTTP response. Specifically, it sends JSON-formatted data including the results of suggesting mailing to "Department C" to the terminal. The terminal receives and analyzes this data.

[0098] Step 6:

[0099] The user checks the suggestions and selects the appropriate destination.

[0100] The user checks the proposal results displayed on the terminal. For example, if the proposal is to send the item to "Division C," the user can accept the proposal or select another destination. The destination selected by the user is sent from the terminal to the server and recorded in the system. This will be used as reference data in the future.

[0101] (Application example 1)

[0102] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0103] The conventional method of sorting mail and parcels with unknown addresses relies on the experience of the person in charge and on sorting lists, which results in low efficiency and a high risk of misdelivery. In addition, there is a problem of reduced accuracy in sorting work because there are insufficient suggestions for the delivery destination based on the sender and the type, size, and format of the delivery.

[0104] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means. In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery item, a means for suggesting possible destinations based on past destinations and detailed information of the delivery item, a means for scanning the delivery item using an image processing device and recognizing the sender and the type and size of the delivery item, and a means for suggesting optimal possible destinations in real time using a machine learning model. This makes it possible to streamline the sorting of mail and parcels with unknown destinations, reduce the risk of misdelivery, and improve the accuracy of the sorting work.

[0105] "Undelivered mail or parcel" refers to mail or parcel whose delivery destination cannot be identified using normal delivery procedures because the recipient's address or name is unclear.

[0106] "Sorting" refers to the task of classifying mail and parcels according to their destination and content, and then distributing them to the appropriate destination.

[0107] A "sender" is an individual or organization that sends mail or parcels.

[0108] "Type of delivery" refers to the classification of the contents of a mail or parcel, and includes different categories such as documents, merchandise, parts, etc.

[0109] "Item size" refers to the physical size or dimensions of a mail or parcel.

[0110] "Form of delivery" refers to the shape and type of packaging of mail or parcels.

[0111] "Candidate destinations" are delivery destinations suggested for unknown mail or parcels based on past data and current information.

[0112] An "image processing device" is a device that captures images using a camera or sensor, analyzes the images, and extracts the necessary information.

[0113] A "machine learning model" is an algorithm that learns patterns based on past data and makes predictions and classifications for new data.

[0114] "Real-time" refers to a state in which the time from data acquisition to processing and output of results is extremely short, with results being obtained almost instantly.

[0115] As an embodiment of the present invention, a system for streamlining the sorting of mail and parcels with unknown destinations at a logistics center will be described. In this system, workers use smart glasses to scan deliveries and suggest optimal drop-off destinations in real time.

[0116] Hardware and Software Used

[0117] Hardware:

[0118] Smart glasses (e.g. Google® Glass®)

[0119] Camera (built into smart glasses)

[0120] software:

[0121] OpenCV (image processing library)

[0122] scikit-learn (machine learning library)

[0123] SQLite (database)

[0124] System Configuration

[0125] 1. Image processing device:

[0126] The smart glasses use a built-in camera to scan the package, and the images captured by the camera are processed using OpenCV to identify the sender and the type and size of the package.

[0127] 2. Database:

[0128] SQLite is used to store detailed information about past mailing destinations and deliveries, which makes it possible to suggest mailing destination candidates based on past data.

[0129] 3. Machine learning models:

[0130] Using scikit-learn, we train a KNN (k-Nearest Neighbors) model based on historical data, which then suggests optimal delivery locations in real time based on the sender and the type, size, and format of the parcel.

[0131] Processing flow

[0132] 1. Scan:

[0133] A worker wears smart glasses and scans the package with a camera. The image captured by the camera is processed using OpenCV to identify the sender and the type and size of the package.

[0134] 2. Database Matching:

[0135] The recognized information is checked against a SQLite database to obtain details about past mailing destinations and deliveries.

[0136] 3. Suggestions for possible recipients:

[0137] Using a machine learning model (KNN), the system will suggest the best possible recipient in real time based on the recognized information and past data, and this information will be displayed on the smart glasses' display.

[0138] Examples

[0139] For example, if the sender is "Company A," the delivery item is "Parts Type B" and the size is "Medium," the system will suggest that the item be sent to "Department C" based on past data. This suggestion will be displayed in real time on the smart glasses' display.

[0140] Examples of prompt statements

[0141] "Please suggest the best delivery location for the item whose sender is Company A, the item is a type B part, and the size is medium."

[0142] In this way, the sorting of mail and parcels with unknown destinations can be made more efficient, the risk of misdelivery can be reduced, and the accuracy of the sorting process can be improved.

[0143] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0144] Step 1:

[0145] A user wears smart glasses and scans a package with the camera. The input is an image of the package captured by the camera, and the output is image data. The camera of the smart glasses captures the image of the package and performs image processing using OpenCV. Specifically, the sender, type of package, and size are extracted from the image.

[0146] Step 2:

[0147] The server uses an image processing device to recognize the sender and the type and size of the delivery from the scanned image. The input is the image data acquired in step 1, and the output is the recognized sender, type and size of the delivery. OpenCV is used to analyze the image and extract text information and dimensions.

[0148] Step 3:

[0149] The server uses the information it has learned to match it with a SQLite database. The input is the sender, type of delivery, and size of the delivery, and the output is details of past delivery destinations and deliveries. A database query is performed to obtain details of past delivery destinations and deliveries.

[0150] Step 4:

[0151] The server uses a machine learning model (KNN) to suggest optimal delivery destination candidates in real time. The input is the sender, the type of delivery, the size, and past delivery destination information, and the output is the optimal delivery destination candidate. The KNN model is run using scikit-learn to predict the optimal delivery destination.

[0152] Step 5:

[0153] The server displays the suggested drop-off candidates on the display of the smart glasses. The input is the optimal drop-off candidate, and the output is the information to be displayed on the display of the smart glasses. The drop-off candidates are displayed on the display of the smart glasses in real time and notified to the user.

[0154] In this way, the sorting of mail and parcels with unknown destinations can be made more efficient, the risk of misdelivery can be reduced, and the accuracy of the sorting process can be improved.

[0155] Example 2

[0156] Next, a second embodiment of the second embodiment will be described. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0157] Sorting mail and parcels with unknown addresses often depends on the work experience and sorting lists of the person in charge, which can be inefficient. In addition, there is a risk that the mail will be delivered to the wrong address because the details of past mailing destinations and deliveries cannot be properly utilized. Furthermore, there is an issue that it is difficult to suggest possible mailing destinations based on the sender and the type, size, and format of the delivery.

[0158] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0159] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for generating possible destinations using a generative AI model. This makes it possible to efficiently sort mail and parcels with unknown addresses and prevent delivery to the wrong destination.

[0160] "Sender" refers to the person or organization sending the mail or package.

[0161] "Delivery" refers to goods or documents sent by mail or courier.

[0162] "Type" indicates the classification of the delivery item, and refers to categories such as documents, parts, and goods.

[0163] "Size" refers to the size of the delivery item, such as small, medium, large, etc.

[0164] "Form" refers to the shape of the delivery item and the state of packaging, such as a box, envelope, or pallet.

[0165] "Potential drop-off locations" refer to locations or departments where a delivery may be dropped off.

[0166] "Previous delivery destination" refers to a location or department where a delivery has previously been delivered.

[0167] "Detailed information" refers to specific data and historical information about the delivery.

[0168] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to analyze data and generate appropriate recipient candidates.

[0169] "Database" refers to a system for storing and managing past mailing history and detailed delivery information.

[0170] "User" refers to the person or organization that operates the system to input data and review / decide on suggested recipients.

[0171] The present invention is a system for streamlining the sorting of mail and parcels with unknown addresses. This system is equipped with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. A specific embodiment of this system will be described below.

[0172] Hardware and Software Used

[0173] Hardware: Servers, terminals (PCs, tablets, etc.)

[0174] Software: Generative AI models (e.g. GPT-4®), database management systems (e.g. MySQL®), postal and parcel management systems

[0175] System program processing

[0176] 1. Data Entry

[0177] The user uses the terminal to enter data such as sender information and the type, size and format of the delivery item.

[0178] Example: A user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal and clicks the send button.

[0179] 2. Database Reference

[0180] The server uses a database management system (e.g. MySQL) to look up details of past mailings and deliveries.

[0181] Example: The server executes an SQL query and sends the following command to the database: "SELECT FROM mailing history WHERE sender = 'Company A' AND delivery = 'Parts of type B'".

[0182] 3. Generating recipient candidates

[0183] The server uses a generative AI model (e.g. GPT-4) to generate potential recipients based on the input data and past data.

[0184] Example: The server sends a prompt to the generative AI model: "If the sender is Company A and the delivery is a type B part, please suggest an appropriate delivery destination based on past data." The generative AI model responds: "Department C is appropriate."

[0185] 4. Suggestions for possible recipients

[0186] The server transmits the generated candidate mailbox destinations to the terminal and suggests them to the user.

[0187] Example: The server sends a message to the terminal saying, "We suggest posting to Department C." The content of the suggestion is displayed on the terminal screen.

[0188] 5. User confirmation and final decision

[0189] The user checks the proposed destination candidates and decides on the final destination.

[0190] Example: The user checks the proposal displayed on the terminal screen and clicks the "Approve posting to Department C" button. The user's selection is sent to the server, and the final posting destination is confirmed.

[0191] Examples of prompt statements

[0192] If the sender is "Company A", the delivery item is "Type B Parts" and the size is "Medium", please suggest an appropriate drop-off location based on past data.

[0193] In this way, users can efficiently sort mail and parcels that are not addressed to their destination. By utilizing generative AI models and a database management system, the system can accurately suggest possible recipients based on past data.

[0194] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0195] Step 1:

[0196] Data Entry

[0197] The user uses the terminal to enter data such as sender information and the type, size and format of the delivery item.

[0198] Input: sender information (e.g. Company A), type of delivery (e.g. Type B parts), size (e.g. medium size)

[0199] Specific operation: The user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal and clicks the send button.

[0200] Output: The entered data is sent to the server.

[0201] Step 2:

[0202] Database Reference

[0203] The server uses a database management system (e.g. MySQL) to look up details of past mailings and deliveries.

[0204] Input: Data entered by the user (sender information, type of delivery, size)

[0205] Specific operation: The server executes an SQL query and sends the command "SELECT FROM mailing history WHERE sender = 'Company A' AND delivery = 'Parts of type B'" to the database.

[0206] Output: Detailed information about past mailing destinations and deliveries is obtained.

[0207] Step 3:

[0208] Generating recipient candidates

[0209] The server uses a generative AI model (e.g. GPT-4) to generate potential recipients based on the input data and past data.

[0210] Input: Data entered by the user, details of past mailings and deliveries retrieved from a database

[0211] Specific operation: The server sends a prompt to the generative AI model: "If the sender is Company A and the delivery is a type B part, please suggest an appropriate delivery destination based on past data." The generative AI model responds: "Department C is appropriate."

[0212] Output: A list of possible recipients (e.g. Department C) is generated.

[0213] Step 4:

[0214] Suggestions for possible recipients

[0215] The server transmits the generated candidate mailbox destinations to the terminal and suggests them to the user.

[0216] Input: Generated recipient candidates (e.g. Department C)

[0217] Specific operation: The server sends a message to the terminal saying, "We suggest posting to Department C." The content of the suggestion is displayed on the terminal screen.

[0218] Output: The suggestions are displayed to the user.

[0219] Step 5:

[0220] User confirmation and final decision

[0221] The user checks the proposed destination candidates and decides on the final destination.

[0222] Input: Suggested recipient (e.g. Department C)

[0223] Specific operation: The user checks the proposal displayed on the terminal screen and clicks the "Approve submission to Department C" button. The user's selection is sent to the server.

[0224] Output: The final posting destination is confirmed.

[0225] In this way, users can efficiently sort mail and parcels that are not addressed to their destination. By utilizing generative AI models and a database management system, the system can accurately suggest possible recipients based on past data.

[0226] (Application example 2)

[0227] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0228] The conventional method of sorting mail and parcels with unknown addresses requires the worker to check the work experience and sorting list each time, which is inefficient and prone to human error. In addition, there is no system that uses past data to suggest the optimal destination, so there is a need to improve the accuracy and efficiency of the sorting process.

[0229] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0230] In this invention, the server includes a means for suggesting possible post destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible post destinations based on past post destinations and detailed information on the delivery, a means installed in the smartphone for suggesting the optimal post destination based on past data, and a means for saving the history of the suggested post destinations and improving the accuracy of the suggestions from the next time onwards. This makes it possible to significantly improve the efficiency and accuracy of the sorting work for mail and parcels with unknown addresses.

[0231] "Unknown address mail or parcel" refers to mail or parcel that cannot be delivered to a specific address through normal delivery procedures because the address specified by the sender is unclear.

[0232] "Sorting work" refers to the task of classifying mail and parcels based on their destination and content, and then distributing them to the appropriate destination.

[0233] "Sender" refers to the individual or organization sending the mail or parcel.

[0234] "Type of delivery" refers to the category of the contents of a mail or parcel, such as documents, merchandise, parts, etc.

[0235] "Size" refers to the physical size of a mail item or parcel, and is classified as small, medium, large, etc.

[0236] "Format" refers to the shape and type of packaging of a mail or parcel, such as an envelope, box, or pallet.

[0237] "Potential drop-off locations" refer to locations or departments to which undeliverable mail or parcels may be delivered.

[0238] "Previous mailing destination" refers to a location or department to which undeliverable mail or parcels have previously been delivered.

[0239] "Detailed information" refers to specific data and history about mail or parcels.

[0240] A "smartphone" is a type of mobile phone and a multi-function device that can connect to the Internet and install applications.

[0241] "Saving history" means recording information about past mailing destinations and deliveries in a database so that it can be referenced later.

[0242] "Suggestion accuracy" refers to the system's ability to accurately suggest the optimal drop-off destination.

[0243] As an embodiment of the present invention, a system for streamlining the sorting of mail and parcels with unknown destinations is constructed. A specific embodiment of the system is described below.

[0244] System Configuration

[0245] Hardware:

[0246] Smartphone (iOS or ANDROID (registered trademark))

[0247] software:

[0248] Python

[0249] Pandas (Data Processing)

[0250] Scikit-learn (machine learning model)

[0251] Program processing explanation

[0252] Data Loading:

[0253] The server reads past delivery data from a CSV file, which gives information about the sender, type of delivery, size, format, and past delivery destination.

[0254] Split features and labels:

[0255] Based on the data it reads, the server classifies the sender, type of delivery, size, and format as features, and the delivery destination as a label.

[0256] Model training:

[0257] The server trains a machine learning model using K-nearest neighbors (KNN) to improve its ability to suggest optimal drop destinations based on past data.

[0258] Suggested destinations:

[0259] When a user uses their smartphone to input information about a new delivery (sender, type of delivery, size, format), the server uses the trained model to suggest the optimal delivery location.

[0260] Save history:

[0261] The server stores the history of suggested destinations in a database, which improves the accuracy of suggestions from the next time onwards.

[0262] Examples

[0263] For example, if the sender is "Company A," the delivery item is a "Type B part," the size is "medium," and the format is "box," it will suggest posting it to "Department C" based on past data.

[0264] Examples of prompt statements

[0265] Please use the following information to suggest the best place to send your message.

[0266] From: Company A

[0267] Delivery type: B-type parts

[0268] Size: Medium

[0269] Format: Box

[0270] In this way, a system can be realized that improves the efficiency and accuracy of sorting mail and parcels with unknown destinations.

[0271] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0272] Step 1:

[0273] Data Loading

[0274] The server reads past delivery data from a CSV file. The path of the CSV file is specified as input. The server reads the CSV file in data frame format using the Pandas library and obtains information on the sender, type of delivery, size, format, and past recipients. A data frame is generated as output.

[0275] Step 2:

[0276] Splitting features and labels

[0277] Based on the data read, the server separates the sender, type, size, and format of the delivery as features, and the recipient as a label. The data frame generated in step 1 is used as input. The server splits the features and labels into separate data frames, in preparation for training the machine learning model. As output, a feature data frame and a label data frame are generated.

[0278] Step 3:

[0279] Model Training

[0280] The server trains a machine learning model using K-nearest neighbors (KNN). As input, the feature and label dataframes generated in step 2 are used. The server trains the KNN model using the Scikit-learn library to improve its ability to suggest optimal drop destinations based on past data. As output, a trained KNN model is generated.

[0281] Step 4:

[0282] Suggested destination

[0283] A user inputs new parcel information (sender, parcel type, size, format) using a smartphone. The parcel information entered by the user is used as input. The server uses the trained KNN model to suggest the optimal delivery destination based on the input information. As output, the suggested delivery destination is displayed to the user.

[0284] Step 5:

[0285] History storage

[0286] The server saves the history of suggested destinations in a database. As input, the proposed destinations in step 4 and the delivery information entered by the user are used. The server records this information in the database to improve the accuracy of future suggestions. As output, the history data is saved in the database.

[0287] In this way, a system can be realized that improves the efficiency and accuracy of sorting mail and parcels with unknown destinations.

[0288] Furthermore, an emotion engine that estimates the emotion of the user may be combined. That is, the identification processing unit 290 may estimate the emotion of the user using the emotion identification model 59, and perform identification processing using the emotion of the user.

[0289] "Example 1"

[0290] As an embodiment of the present invention, a system incorporating an emotion engine is provided. The system recognizes the user's emotion and adjusts the candidate destinations based on the emotion. Specifically, if the system recognizes that the user is stressed, the system reduces the number of candidate destinations and simplifies the options. Also, if the system recognizes that the user is relaxed, the system provides more candidate destinations, giving the user more freedom of choice. Furthermore, the emotion engine also adjusts the way in which the candidate destinations are presented based on the user's emotion. For example, if the system recognizes that the user is anxious, the system does not display a list of candidate destinations, but directly presents the most likely destination. This allows the user to make a decision quickly.

[0291] "Example 2"

[0292] As an embodiment of the present invention, a system incorporating an emotion engine is provided. The system recognizes the user's emotion and adjusts the candidate destinations based on the emotion. Specifically, if the system recognizes that the user is stressed, the system reduces the number of candidate destinations and simplifies the options. Also, if the system recognizes that the user is relaxed, the system provides more candidate destinations, giving the user more freedom of choice. Furthermore, the emotion engine also adjusts the way in which the candidate destinations are presented based on the user's emotion. For example, if the system recognizes that the user is anxious, the system does not display a list of candidate destinations, but directly presents the most likely destination. This allows the user to make a decision quickly.

[0293] The process flow of each embodiment will be described below.

[0294] "Example 1"

[0295] Step 1: The system recognizes the user's emotions using an emotion engine, which infers emotions from the user's facial expressions, tone of voice, choice of words, etc.

[0296] Step 2: After the emotion engine recognizes the user's emotion, the system adjusts the candidate destinations based on the emotion. For example, if the system recognizes that the user is stressed, the system reduces the number of candidate destinations to simplify the options. The server's adjustment means may generate the most likely candidate destination as the candidate destination when the user is stressed, and generate multiple candidate destinations when the user is relaxed.

[0297] Step 3: On the other hand, if the system detects that the user is relaxed, it will provide more possible destinations, giving the user more freedom of choice.

[0298] Step 4: Furthermore, the emotion engine will also adjust the way in which the system presents the possible destinations based on the user's emotions. For example, if the system detects that the user is in a hurry, it will directly present the most likely destination instead of displaying a list of possible destinations, allowing the user to make a decision quickly.

[0299] Example 1

[0300] Next, a description will be given of Example 1 of the embodiment 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal".

[0301] Conventional methods for sorting mail and parcels with unknown addresses are inefficient because they require the person in charge to check the work experience and sorting list each time. In addition, there is a problem that appropriate support cannot be provided to users who feel stressed or impatient because the system cannot respond flexibly to the user's emotional state. This has led to a demand for more efficient sorting work and improved user satisfaction.

[0302] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0303] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on the emotions. This makes it possible to streamline the sorting of mail and parcels with unknown addresses and to respond flexibly to the user's emotional state.

[0304] "Sender" refers to the individual or organization sending the mail or parcel.

[0305] "Delivery" refers to goods or documents sent by mail or courier.

[0306] "Type" is an attribute that indicates the classification of the delivery, and includes, for example, documents, parts, and merchandise.

[0307] "Size" is an attribute that indicates the physical size of a delivery item, and includes, for example, length, width, height, weight, and the like.

[0308] "Form" is an attribute that indicates the shape of the delivery item or the state of packaging, and includes, for example, a box, an envelope, a pallet, and the like.

[0309] "Potential drop-off locations" refers to locations or departments where deliveries may be dropped off.

[0310] "Previous mailbox locations" refers to records of locations or departments where deliveries have been previously mailed.

[0311] "Detailed information" refers to specific data and history about the delivery.

[0312] "Emotion recognition" refers to identifying a user's emotional state by analyzing their facial expressions and tone of voice.

[0313] "Adjust" refers to changing the system's behavior or suggestions to suit specific conditions or circumstances.

[0314] The present invention is a system for streamlining the sorting of mail and parcels with unknown destinations. This system includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on those emotions.

[0315] Hardware and Software Used

[0316] Hardware

[0317] Server: Analyzes and stores data.

[0318] Terminal: Provides a user interface and is equipped with a camera and microphone for recognizing the user's emotions.

[0319] Scanner: Scan postal and courier labels.

[0320] software

[0321] Google Cloud Vision API: Converts paper data into digital data using OCR (optical character recognition) technology.

[0322] MySQL: A database management system that stores analysis results and past submission data.

[0323] Microsoft® Azure® Emotion API: Recognizes emotions by analyzing the user's facial expressions and tone of voice.

[0324] System Operation

[0325] Data Entry and Analysis

[0326] The server receives data such as sender information for mail and parcels, and the type, size, and format of the delivery. Specifically, it scans the mail and parcel labels and converts them from paper to digital data using Google Cloud Vision API. The server analyzes the received data and identifies the sender and the type, size, and format of the delivery. The analysis results are stored in a MySQL database.

[0327] Suggestions for possible recipients

[0328] The server searches a database of past delivery destinations based on the analysis results and suggests the most appropriate delivery destination. For example, if the sender is "Company A" and the delivery item is "Type B parts," it suggests delivery to "Department C" based on past data. Also, if the size of the delivery item falls within a certain range, it suggests delivery to a specific location.

[0329] Emotion engine adjustment

[0330] The device uses a camera and microphone to analyze the user's facial expressions and tone of voice to recognize the user's emotions. It uses Microsoft Azure's Emotion API. If the server recognizes that the user is stressed, it reduces the number of possible delivery candidates and simplifies the options. Conversely, if the server recognizes that the user is relaxed, it provides more possible delivery candidates.

[0331] Suggestion of possible recipients

[0332] The server sends the adjusted drop-off candidates to the terminal and presents them to the user. If the server detects that the user is in a hurry, it will directly present the most likely drop-off destination, allowing the user to make a decision quickly.

[0333] Examples of concrete examples and prompts

[0334] Examples

[0335] When a user receives a "Part B" from "Company A", the system will suggest dropping it off at "Division C" based on past data. If the system detects that the user is stressed, it will only suggest dropping it off at "Division C".

[0336] Examples of prompt statements

[0337] "We have received part B from company A. Please suggest the best delivery destination based on past data. Also, please adjust the suggestions based on user sentiment."

[0338] The above is an embodiment of the present invention.

[0339] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0340] Step 1:

[0341] Data Entry

[0342] The server receives data such as sender information for mail or parcels, and the type, size, and format of the delivery. As input, it scans the label of the mail or parcel to obtain image data. In concrete terms, the server obtains image data from the scanner and calls the Google Cloud Vision API to extract text information from the image data. As output, it obtains the extracted text information.

[0343] Step 2:

[0344] 4. Data Analysis

[0345] The server analyzes the text information obtained in step 1 to identify the sender, type of delivery, size, and format. As input, it receives the extracted text information. In concrete terms, the server analyzes the text information to identify the sender, type of delivery, size, and format. As output, it obtains the analysis results, which are stored in a MySQL database.

[0346] Step 3:

[0347] Suggestions for possible recipients

[0348] The server searches the database of past destinations based on the analysis results obtained in step 2, and proposes the most suitable destination candidates. It receives the analysis results as input. In concrete terms, the server queries the MySQL database to retrieve the past destination data. As output, it obtains a list of optimal destination candidates.

[0349] Step 4:

[0350] Emotion engine adjustment

[0351] To recognize the user's emotions, the device uses a camera and microphone to analyze the user's facial expressions and tone of voice. It receives the user's facial expression and voice data as input. Specifically, the device calls Microsoft Azure's Emotion API to analyze the user's emotions. The output is the user's emotional state. The server adjusts the number of possible recipients based on the user's emotional state.

[0352] Step 5:

[0353] Suggestion of possible recipients

[0354] The server sends the adjusted list of destination candidates in step 4 to the terminal and presents them to the user. As input, it receives the adjusted list of destination candidates. In concrete terms, the server sends the adjusted destination candidates to the terminal, and the terminal displays the destination candidates on the display. As output, the destination candidates presented to the user are obtained.

[0355] (Application example 1)

[0356] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0357] The conventional method for sorting mail and parcels with unknown addresses requires the person in charge to check the work experience and sorting list each time, which makes it inefficient and prone to errors. In addition, there are many situations that require flexible responses based on the user's emotions, and there was no system that could handle this.

[0358] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0359] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the number of possible destinations and the presentation method based on the emotions. This makes it possible to streamline the work of sorting mail and parcels with unknown addresses and to respond flexibly to the user's emotions.

[0360] "Sender" refers to the person or organization sending the mail or package.

[0361] "Delivery" refers to goods or documents sent by mail or courier.

[0362] "Type" is an attribute that indicates the classification of the delivery, such as documents, parts, or merchandise.

[0363] "Size" is an attribute that indicates the size or dimensions of the delivery item.

[0364] "Form" is an attribute that indicates the shape of the delivery and the state of packaging.

[0365] "Possible drop-off locations" refers to possible locations where deliveries may be dropped off or potential recipients.

[0366] "Previous delivery destination" refers to a location or recipient where a delivery has previously been delivered.

[0367] "Detailed information" refers to specific data and historical information about the delivery.

[0368] "Emotion recognition" refers to detecting and understanding a user's emotional state.

[0369] "Emotion-based adjustment" refers to changing the system's behavior or presentation in response to recognized emotions.

[0370] A system for implementing the present invention includes a server, a smartphone, an emotion recognition engine, and a machine learning model. Specific embodiments are described below.

[0371] System Configuration

[0372] Hardware

[0373] Server: Stores and processes data.

[0374] Smartphones: Used as a user interface and receive user input.

[0375] Emotion recognition engine: Recognize user emotions in real time.

[0376] Machine learning model: A model for suggesting possible recipients (e.g., KNN model).

[0377] software

[0378] Python: Used to implement the program.

[0379] scikit-learn: Used to train and predict machine learning models.

[0380] emotion_recognition: A library of emotion recognition engines.

[0381] Data processing and calculation

[0382] server

[0383] The server stores past delivery destination data and delivery details, retrieves them when needed, and also trains machine learning models to suggest delivery destinations based on user input.

[0384] Smartphone

[0385] The smartphone receives input from the user (sender, type, size, and format of the delivery item) and sends it to the server. It also recognizes the user's emotions in real time through an emotion recognition engine and sends that information to the server.

[0386] Emotion Recognition Engine

[0387] The emotion recognition engine recognizes the user's emotions from their facial expressions and voice, and returns the results to the smartphone, which then adjusts the number of possible recipients and the way they are presented according to the user's emotional state.

[0388] Machine learning models

[0389] A machine learning model (e.g., KNN model) is trained based on past data and suggests optimal delivery candidates based on user input data. It receives information from an emotion recognition engine and adjusts the suggestions according to the user's emotions.

[0390] Examples

[0391] For example, if the sender is "Company A," the delivery is "Type B parts," and the size is "medium," the system will suggest dropping it off at "Department C" based on past data. If the user is feeling stressed, the system will narrow down the suggestions to one, and if the user is relaxed, it will display up to five suggestions.

[0392] Examples of prompt statements

[0393] Suggest possible delivery destinations for a "medium" size package with a "Type B part" from "Company A". If the user is stressed, narrow down the suggestions to one, and if the user is relaxed, display up to five suggestions.

[0394] In this way, a system is provided that can streamline the sorting of mail and parcels with unknown destinations and can flexibly respond to the user's emotions.

[0395] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0396] Step 1:

[0397] The user uses a smartphone to input the sender, type of delivery, size, and format. The input data is sent from the smartphone to the server. The input data includes the sender name, type of delivery, size, and format.

[0398] Step 2:

[0399] The smartphone uses an emotion recognition engine to recognize the user's emotions in real time. The emotion recognition engine analyzes the user's facial expressions and voice data to detect their emotional state (e.g., stress, relaxation). The detected emotion data is sent from the smartphone to a server.

[0400] Step 3:

[0401] The server retrieves relevant data from a database of past mailing destinations based on the received input data (sender, type of delivery, size, format) and emotion data. The server compares the past mailing destination data with the detailed delivery information to generate optimal mailing destination candidates.

[0402] Step 4:

[0403] The server uses a machine learning model (e.g., KNN model) to suggest possible destinations based on the input data. The machine learning model is trained based on past data and analyzes the input data to predict the best possible destinations. The predicted destination candidates are sent from the server to the smartphone.

[0404] Step 5:

[0405] The server adjusts the number and presentation of possible recipients based on the emotion data. For example, if the user is stressed, the server reduces the number of possible recipients and presents only one most likely recipient. If the user is relaxed, the server presents multiple possible recipients (e.g., five).

[0406] Step 6:

[0407] The smartphone displays the candidate destinations received from the server to the user. The user selects the most suitable destination from the presented candidates. The selected destination is sent from the smartphone to the server.

[0408] Step 7:

[0409] The server records the destination selected by the user and updates the database to improve the accuracy of the suggestions for the next time, allowing the system to continuously learn and improve the accuracy of the suggestions.

[0410] In this way, a system is provided that can streamline the sorting of mail and parcels with unknown destinations and can flexibly respond to the user's emotions.

[0411] Example 2

[0412] Next, a second embodiment of the second embodiment will be described. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0413] Conventional methods for sorting mail and parcels with unknown addresses rely on the work experience and sorting lists of the person in charge, which makes them inefficient and prone to errors. In addition, there is also the problem that the system is unable to respond flexibly to the user's emotional state, making it difficult to provide appropriate support to users who feel stressed or impatient. Furthermore, there are insufficient suggestions for delivery destinations based on the sender and the type, size, and format of the delivery, making it difficult to quickly find the optimal delivery destination.

[0414] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0415] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on the emotions. This makes it possible to streamline the sorting of mail and parcels with unknown destinations and reduce errors. In addition, since the server can flexibly respond to the user's emotional state, it reduces the user's stress and impatience and allows the user to quickly find the optimal destination.

[0416] "Sender" refers to the person or organization sending the mail or package.

[0417] "Delivery" refers to goods or documents sent by mail or courier.

[0418] "Type" refers to the classification or category of the delivery, such as documents, parts, merchandise, etc.

[0419] "Size" refers to the size or dimensions of the delivery item, and includes, for example, small, medium, large, etc.

[0420] "Format" refers to the shape or packaging of the delivery, such as a box, envelope, or pallet.

[0421] "Potential delivery destinations" refers to potential locations or departments where deliveries may be dropped off.

[0422] "Previous delivery destination" refers to a location or department where a delivery has previously been delivered.

[0423] "Detailed information" refers to specific data and historical information about the delivery.

[0424] "Emotion recognition engine" refers to software or hardware for analyzing and recognizing a user's emotional state.

[0425] A "generative AI model" refers to a model that uses artificial intelligence to analyze data and generate optimal results.

[0426] A "prompt" refers to an instruction or question that is input into a generative AI model.

[0427] The present invention is a system for streamlining the sorting of mail and parcels with unknown addresses, and includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery item, a means for suggesting possible destinations based on past delivery destinations and detailed information about the delivery item, and a means for recognizing the user's emotions and adjusting the possible destinations based on those emotions.

[0428] Hardware and Software Used

[0429] Hardware: Servers, user devices (PCs, tablets, smartphones)

[0430] Software: Generative AI models, emotion recognition engine, database management system (DBMS)

[0431] Specific operation of the system

[0432] 1. Data Entry:

[0433] The user uses the terminal to enter information such as the sender and the type, size, and format of the delivery.

[0434] Specific operation: The user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal.

[0435] 2. Data Analysis:

[0436] The server receives the entered data and compares it with its historical database.

[0437] Specific operation: The server uses a database management system (DBMS) to search for data that "Parts of type B from Company A were previously mailed to Department C."

[0438] 3. Suggestions for possible recipients:

[0439] The server uses a generative AI model to suggest optimal delivery candidates.

[0440] Specific operation: The server provides input data as prompts to the generative AI model to generate optimal delivery candidates.

[0441] Example: The server inputs the prompt "Sender: Company A, Delivery: Type B part, Size: Medium" into the generative AI model and outputs "Department C" as a possible recipient.

[0442] 4. Emotion recognition:

[0443] The terminal uses an emotion recognition engine to analyze the user's emotions in real time while the user is operating the terminal.

[0444] Specific behavior: Uses the device's camera and microphone to analyze the user's facial expressions and tone of voice to determine their emotional state.

[0445] Example: The device captures the user's face with a camera and an emotion recognition engine detects "stress."

[0446] 5. Adjustment of possible recipients:

[0447] The server adjusts the number of possible recipients and the presentation method based on the user's feelings.

[0448] What it does: The server receives data from the emotion recognition engine and refines the list of potential recipients.

[0449] Example: The server receives information that "the user is feeling stressed" and narrows down the possible recipients to only "Department C".

[0450] Examples

[0451] Example of input prompt:

[0452] "If the sender is Company A and the delivery item is Type B parts, please suggest the best delivery location based on past data."

[0453] "If users are feeling stressed, present them with fewer possible recipients."

[0454] In this way, the system suggests optimal candidates for recipients based on the user's input data and emotional state.

[0455] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0456] Step 1:

[0457] Data Entry

[0458] The user uses the terminal to enter information such as the sender and the type, size, and format of the delivery.

[0459] Input: sender, delivery type, size, format

[0460] Specific operation: The user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal.

[0461] Output: The entered data is sent to the server.

[0462] Step 2:

[0463] 4. Data Analysis

[0464] The server receives the entered data and compares it with its historical database.

[0465] Input: Data sent by the user, such as sender, type of delivery, size, and format

[0466] Specific operation: The server uses a database management system (DBMS) to search for data that "Parts of type B from Company A were previously mailed to Department C."

[0467] Output: The matching result is the past mailing destination data.

[0468] Step 3:

[0469] Suggestions for possible recipients

[0470] The server uses a generative AI model to suggest optimal delivery candidates.

[0471] Input: Past mailing destination data obtained as a result of matching, and data input by users

[0472] Specific operation: The server provides input data as prompts to the generative AI model to generate optimal delivery candidates.

[0473] Output: A list of potential recipients from the generative AI model.

[0474] Step 4:

[0475] emotion recognition

[0476] The terminal uses an emotion recognition engine to analyze the user's emotions in real time while the user is operating the terminal.

[0477] Input: User facial expressions and tone of voice

[0478] Specific behavior: Uses the device's camera and microphone to analyze the user's facial expressions and tone of voice to determine their emotional state.

[0479] Output: Emotion state data from the emotion recognition engine.

[0480] Step 5:

[0481] Adjustment of possible recipients

[0482] The server adjusts the number of possible recipients and the presentation method based on the user's feelings.

[0483] Input: Emotion state data from an emotion recognition engine, a list of recipient candidates from a generative AI model

[0484] What it does: The server receives data from the emotion recognition engine and refines the list of potential recipients.

[0485] Output: The refined list of possible recipients is presented to the user.

[0486] In this way, the system suggests optimal candidates for recipients based on the user's input data and emotional state.

[0487] (Application example 2)

[0488] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0489] Sorting mail and parcels with unknown addresses is difficult because it is difficult to find the appropriate place to drop them off based on the sender and the type, size, and format of the parcel, and because it requires the staff member to check the work experience and sorting list each time.Furthermore, flexible responses according to the user's emotional state are required, but conventional systems were unable to achieve this.

[0490] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0491] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information of the delivery, a means for recognizing the user's emotions and adjusting the possible destinations based on the emotions, a means for scanning the delivery with a camera in the smart glasses and recognizing the sender and the type, size, and format of the delivery, and a means for presenting the optimal possible destinations based on the recognized information, past data, and the user's emotions. This makes it possible to efficiently and flexibly sort mail and parcels with unknown addresses.

[0492] A "sender" is a person or organization that sends a letter or package.

[0493] A "delivery" is an item sent by mail or courier.

[0494] "Type" is an attribute that indicates the classification of the delivery.

[0495] "Size" is an attribute that indicates the size of the delivery item.

[0496] "Form" is an attribute that indicates the shape of the delivery item and the type of packaging.

[0497] "Potential delivery destinations" are potential locations or departments where deliveries may be delivered.

[0498] A "previous delivery destination" is a location or department where a delivery has been previously delivered.

[0499] "Detailed information" refers to specific data and history information regarding the delivery.

[0500] "User emotion" refers to the psychological state of a person using a system.

[0501] "Smart glasses" are glasses-type devices that have built-in cameras and displays and have the ability to display information.

[0502] A "camera" is a device that captures images or videos.

[0503] "Recognition" refers to the system analyzing input information and determining specific attributes or conditions.

[0504] "Data" is a collection of information that is processed by a system.

[0505] "Presentation" means that the system displays information to the user.

[0506] A system for implementing the present invention includes smart glasses, a camera, an emotion recognition engine, a database, and a display for presenting possible recipients. The specific configuration and operation of the system will be described below.

[0507] System Configuration

[0508] 1. Smart glasses: The smart glasses are equipped with a camera and a display and are worn by workers. The camera scans the package and the display shows possible recipients.

[0509] 2. Camera: The camera built into the smart glasses takes pictures of the package and provides data to recognize the sender, package type, size and format.

[0510] 3. Emotion Recognition Engine: The emotion recognition engine scans the user's face and recognizes their emotions, allowing it to tailor the recipient suggestions to suit the user's psychological state.

[0511] 4. Database: The database stores detailed information on past mailing destinations and deliveries, and uses this data to suggest possible mailing destinations.

[0512] 5. A means for presenting possible recipients: A means for presenting optimal recipient candidates based on recognized information, past data, and the user's emotions.

[0513] System Operation

[0514] 1. Parcel scanning: A worker wears smart glasses and scans the parcel with a camera, which captures an image of the parcel and identifies the sender, type, size and format of the parcel.

[0515] 2. Database query: Based on the recognized information, detailed information on past mailing destinations and deliveries is obtained from the database.

[0516] 3. Emotion Recognition: The camera in the smart glasses scans the user's face and the emotion recognition engine recognizes the user's emotions.

[0517] 4. Suggestion of possible recipients: Suggest optimal recipients based on recognized information, past data, and user emotions. If the user is stressed, the number of possible recipients will be reduced, and if the user is relaxed, more possible recipients will be provided.

[0518] 5. Presentation of possible drop-off destinations: Potential drop-off destinations are displayed on the smart glasses display, allowing the user to make a quick decision.

[0519] Examples

[0520] When a logistics center worker wears smart glasses and scans an unknown delivery, the system automatically recognizes the sender and the type, size, and format of the delivery, and suggests the best possible delivery destination based on past data and the user's emotions. For example, if the sender is "Company A" and the delivery is "Type B parts," the system will suggest delivery to "Department C" based on past data.

[0521] Examples of prompt statements

[0522] "If the sender is Company A and the delivery is Type B parts, suggest the best possible delivery destinations based on past data. Also, if the user is stressed, provide fewer delivery destinations, and if the user is relaxed, provide more delivery destinations."

[0523] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0524] Step 1:

[0525] The user puts on the smart glasses and scans the delivery with the camera.

[0526] Input: Delivery image

[0527] Output: Image data

[0528] Specific operation: The camera in the smart glasses takes a picture of the delivery item and captures the image data.

[0529] Step 2:

[0530] The server analyzes the image data and identifies the sender and the type, size and format of the delivery.

[0531] Input: Image data

[0532] Output: sender, delivery type, size, format information

[0533] What it does: Uses image recognition algorithms to extract sender, package type, size and format from image data.

[0534] Step 3:

[0535] Based on the information recognized by the server, detailed information about past mailing destinations and deliveries is retrieved from the database.

[0536] Input: sender, delivery type, size, and format information

[0537] Output: Details of past mailing destinations and deliveries

[0538] What it does: Executes database queries to retrieve details about past mailings and deliveries.

[0539] Step 4:

[0540] The user scans their face with the smart glasses' camera, and the emotion recognition engine recognizes the user's emotions.

[0541] Input: User's face image

[0542] Output: The user's emotional state.

[0543] What it does: It uses emotion recognition algorithms to analyze the user's emotional state from facial images.

[0544] Step 5:

[0545] The server suggests optimal candidates for delivery destinations based on the information recognized by the server, past data, and user sentiment.

[0546] Input: sender, delivery type, size, format information, past mailing destinations and delivery details, user emotional state

[0547] Output: List of possible recipients

[0548] What it does: It uses a generative AI model to analyze input data and generate optimal delivery candidates.

[0549] Step 6:

[0550] The server displays possible delivery destinations on the smart glasses' display.

[0551] Input: List of possible recipients

[0552] Output: Possible recipients displayed on the smart glasses display

[0553] Specific operation: The list of possible recipients is sent to the smart glasses display and displayed to the user.

[0554] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0555] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0556] Another example of generative AI is Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0557] In the above embodiment, an example was given in which the specific process was performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0558] [Second embodiment]

[0559] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0560] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0561] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a wide area network (WAN) and / or a local area network (LAN).

[0562] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0563] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0564] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0565] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0566] Fig. 4 shows an example of main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0567] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0568] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0569] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0570] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0571] "Example 1"

[0572] The system of the present invention is for sorting mail and parcels with unknown addresses. Specifically, it is provided with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. For example, if the sender is "Company A" and the delivery is "Parts Type B," it is suggested to post the delivery to "Department C" based on past data. In addition, if the size of the delivery falls within a specific range, it is possible to suggest possible destinations according to various conditions, such as suggesting posting to a specific location.

[0573] "Example 2"

[0574] The system of the present invention is for sorting mail and parcels with unknown addresses. Specifically, it is provided with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. For example, if the sender is "Company A" and the delivery is "Parts Type B," it is suggested to post the delivery to "Department C" based on past data. In addition, if the size of the delivery falls within a specific range, it is possible to suggest possible destinations according to various conditions, such as suggesting posting to a specific location.

[0575] The process flow of each embodiment will be described below.

[0576] "Example 1"

[0577] Step 1: The system receives information about the bounced mail or parcel, including the sender, type, size, and format of the parcel.

[0578] Step 2: Next, the system will suggest possible delivery destinations based on the received information. Specifically, it will suggest possible delivery destinations based on the sender and the type, size and format of the delivery.

[0579] Step 3: The system then suggests possible delivery destinations based on past delivery destinations and detailed information about the delivery item. For example, if the sender is "Company A" and the delivery item is "Parts Type B," the system will suggest delivery to "Department C" based on past data.

[0580] Step 4: Finally, the system presents the proposed destination candidates to the user, who can select the most suitable destination from the presented destination candidates.

[0581] Example 1

[0582] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal".

[0583] Sorting mail and parcels with unknown addresses requires the staff to check their work experience and sorting list each time, which is inefficient and increases the risk of misdelivery. There is also the problem that it is difficult to quickly find the appropriate delivery destination based on the sender and the type, size, and format of the delivery. To solve these issues, a system is needed that utilizes past data to efficiently and accurately suggest delivery destination candidates.

[0584] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0585] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, a means for generating possible destinations using a generative AI model, a means for transmitting data entered by a user to the server, a means for the server to acquire past data from a database, a means for the server to transmit the proposed results to a terminal, and a means for the user to check the proposed results and select an appropriate destination. This makes it possible to efficiently and accurately sort mail and parcels with unknown addresses.

[0586] "Sender" refers to the person or organization sending the mail or package.

[0587] "Delivery" refers to items sent by mail or courier service.

[0588] "Type" is an attribute that indicates the classification of the delivery.

[0589] "Size" is an attribute that indicates the size of the delivery item.

[0590] "Form" is an attribute that indicates the shape of the delivery item and the type of packaging.

[0591] "Possible delivery destinations" refer to locations or departments where deliveries may be dropped off.

[0592] "Previous mailbox location" refers to a location or department where a delivery has been previously mailed.

[0593] "Detailed information" refers to specific data and history about the delivery.

[0594] A "generative AI model" is a model that uses artificial intelligence to analyze data and generate potential recipients.

[0595] A "user" is a person who operates the system to enter postal or courier information.

[0596] A "server" is a computer system that processes and stores data.

[0597] A "terminal" is a computer that includes input and display devices operated by a user.

[0598] A "database" is a system for storing detailed information about past mailing destinations and deliveries.

[0599] "Suggested results" refer to the candidate recipients generated by the generative AI model.

[0600] This invention is a system for efficiently and accurately sorting mail and parcels with unknown destinations. This system is realized using a server, a terminal, a generative AI model, and a database management system (DBMS).

[0601] Hardware and Software Used

[0602] Hardware: Servers, terminals

[0603] Software: Generative AI models, Database Management System (DBMS)

[0604] System Configuration

[0605] 1. User enters postal or courier information:

[0606] The user inputs information about the mail or parcel they received, such as the sender, type, size, and format of the parcel, into the input form on the terminal. For example, the sender is "Company A," the parcel is "Type B parts," the size is "medium," and the format is "box."

[0607] 2. The device sends the input data to the server:

[0608] The terminal sends the data entered by the user to the server, specifically, using an HTTP request to send information such as the sender, type of delivery, size, format, etc.

[0609] 3. The server retrieves historical data from the database:

[0610] The server uses a database management system (DBMS) to retrieve past mailing destinations and delivery details. For example, it executes an SQL query to retrieve past mailing destination data for "Parts of type B" from "Company A."

[0611] 4. The server uses the generative AI model to generate possible recipients:

[0612] The server inputs the acquired past data into the generative AI model to generate candidates for recipients. Specifically, it inputs the following prompt sentence to the generative AI model:

[0613] If the sender is "Company A" and the delivery is "Parts Type B", which department would you suggest to post it to based on past data?

[0614] Based on this prompt, the generative AI model suggests sending the item to "Division C."

[0615] 5. The server sends the proposal results to the device:

[0616] The server sends the candidate mailing destinations generated by the generative AI model to the terminal. Specifically, it uses an HTTP response to send the result proposing mailing to "Department C" to the terminal.

[0617] 6. User reviews the suggestions and selects the appropriate destination:

[0618] The user checks the proposed results displayed on the terminal. For example, if the proposal is to send the item to "Division C," the user can accept the proposal or select another destination. The destination selected by the user is recorded in the system and used as reference data for the future.

[0619] In this way, users can efficiently find the correct drop-off location for undelivered mail or parcels, improving the efficiency of sorting operations and reducing the risk of misdelivery.

[0620] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0621] Step 1:

[0622] A user enters postal or courier information.

[0623] The user inputs information about the mail or parcel they received, such as the sender, type of item, size, and format, into the input form on the terminal. For example, the sender is "Company A," the item is "Type B parts," the size is "medium," and the format is "box." The input data is temporarily stored in the terminal's memory.

[0624] Step 2:

[0625] The terminal transmits the input data to the server.

[0626] The terminal sends the data entered by the user to the server using an HTTP request. Specifically, it sends JSON-formatted data including information such as the sender, type of delivery, size, and format to the server. The server receives and analyzes this data.

[0627] Step 3:

[0628] The server retrieves historical data from the database.

[0629] The server uses a database management system (DBMS) to retrieve past mailing destinations and detailed delivery information. For example, it executes an SQL query to retrieve past mailing destination data for "Part B" from "Company A." The retrieved data is stored in the server's memory.

[0630] Step 4:

[0631] The server uses the generative AI model to generate candidate recipients.

[0632] The server inputs the acquired past data into the generative AI model to generate candidates for recipients. Specifically, it inputs the following prompt sentence to the generative AI model:

[0633] If the sender is "Company A" and the delivery is "Parts Type B", which department would you suggest to post it to based on past data?

[0634] Based on this prompt, the generative AI model suggests sending the mail to "Division C." The generated mail destination candidates are stored in the server's memory.

[0635] Step 5:

[0636] The server transmits the proposal results to the terminal.

[0637] The server sends the candidate mailing destinations generated by the generative AI model to the terminal using an HTTP response. Specifically, it sends JSON-formatted data including the results of suggesting mailing to "Department C" to the terminal. The terminal receives and analyzes this data.

[0638] Step 6:

[0639] The user checks the suggestions and selects the appropriate destination.

[0640] The user checks the proposal results displayed on the terminal. For example, if the proposal is to send the item to "Division C," the user can accept the proposal or select another destination. The destination selected by the user is sent from the terminal to the server and recorded in the system. This will be used as reference data in the future.

[0641] (Application example 1)

[0642] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal".

[0643] The conventional method of sorting mail and parcels with unknown addresses relies on the experience of the person in charge and on sorting lists, which results in low efficiency and a high risk of misdelivery. In addition, there is a problem of reduced accuracy in sorting work because there are insufficient suggestions for the delivery destination based on the sender and the type, size, and format of the delivery.

[0644] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means. In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery item, a means for suggesting possible destinations based on past destinations and detailed information of the delivery item, a means for scanning the delivery item using an image processing device and recognizing the sender and the type and size of the delivery item, and a means for suggesting optimal possible destinations in real time using a machine learning model. This makes it possible to streamline the sorting of mail and parcels with unknown destinations, reduce the risk of misdelivery, and improve the accuracy of the sorting work.

[0645] "Undelivered mail or parcel" refers to mail or parcel whose delivery destination cannot be identified using normal delivery procedures because the recipient's address or name is unclear.

[0646] "Sorting" refers to the task of classifying mail and parcels according to their destination and content, and then distributing them to the appropriate destination.

[0647] A "sender" is an individual or organization that sends mail or parcels.

[0648] "Type of delivery" refers to the classification of the contents of a mail or parcel, and includes different categories such as documents, merchandise, parts, etc.

[0649] "Item size" refers to the physical size or dimensions of a mail or parcel.

[0650] "Form of delivery" refers to the shape and type of packaging of mail or parcels.

[0651] "Candidate destinations" are delivery destinations suggested for unknown mail or parcels based on past data and current information.

[0652] An "image processing device" is a device that captures images using a camera or sensor, analyzes the images, and extracts the necessary information.

[0653] A "machine learning model" is an algorithm that learns patterns based on past data and makes predictions and classifications for new data.

[0654] "Real-time" refers to a state in which the time from data acquisition to processing and output of results is extremely short, with results being obtained almost instantly.

[0655] As an embodiment of the present invention, a system for streamlining the sorting of mail and parcels with unknown destinations at a logistics center will be described. In this system, workers use smart glasses to scan deliveries and suggest optimal drop-off destinations in real time.

[0656] Hardware and Software Used

[0657] Hardware:

[0658] Smart glasses (e.g. Google Glass)

[0659] Camera (built into smart glasses)

[0660] software:

[0661] OpenCV (image processing library)

[0662] scikit-learn (machine learning library)

[0663] SQLite (database)

[0664] System Configuration

[0665] 1. Image processing device:

[0666] The smart glasses use a built-in camera to scan the package, and the images captured by the camera are processed using OpenCV to identify the sender and the type and size of the package.

[0667] 2. Database:

[0668] SQLite is used to store detailed information about past mailing destinations and deliveries, which makes it possible to suggest mailing destination candidates based on past data.

[0669] 3. Machine learning models:

[0670] Using scikit-learn, we train a KNN (k-Nearest Neighbors) model based on historical data, which then suggests optimal delivery locations in real time based on the sender and the type, size, and format of the parcel.

[0671] Processing flow

[0672] 1. Scan:

[0673] A worker wears smart glasses and scans the package with a camera. The image captured by the camera is processed using OpenCV to identify the sender and the type and size of the package.

[0674] 2. Database Matching:

[0675] The recognized information is checked against a SQLite database to obtain details about past mailing destinations and deliveries.

[0676] 3. Suggestions for possible recipients:

[0677] Using a machine learning model (KNN), the system will suggest the best possible recipient in real time based on the recognized information and past data, and this information will be displayed on the smart glasses' display.

[0678] Examples

[0679] For example, if the sender is "Company A," the delivery item is "Parts Type B" and the size is "Medium," the system will suggest that the item be sent to "Department C" based on past data. This suggestion will be displayed in real time on the smart glasses' display.

[0680] Examples of prompt statements

[0681] "Please suggest the best delivery location for the item whose sender is Company A, the item is a type B part, and the size is medium."

[0682] In this way, the sorting of mail and parcels with unknown destinations can be made more efficient, the risk of misdelivery can be reduced, and the accuracy of the sorting process can be improved.

[0683] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0684] Step 1:

[0685] A user wears smart glasses and scans a package with the camera. The input is an image of the package captured by the camera, and the output is image data. The camera of the smart glasses captures the image of the package and performs image processing using OpenCV. Specifically, the sender, type of package, and size are extracted from the image.

[0686] Step 2:

[0687] The server uses an image processing device to recognize the sender and the type and size of the delivery from the scanned image. The input is the image data acquired in step 1, and the output is the recognized sender, type and size of the delivery. OpenCV is used to analyze the image and extract text information and dimensions.

[0688] Step 3:

[0689] The server uses the information it has learned to match it with a SQLite database. The input is the sender, type of delivery, and size of the delivery, and the output is details of past delivery destinations and deliveries. A database query is performed to obtain details of past delivery destinations and deliveries.

[0690] Step 4:

[0691] The server uses a machine learning model (KNN) to suggest optimal delivery destination candidates in real time. The input is the sender, the type of delivery, the size, and past delivery destination information, and the output is the optimal delivery destination candidate. The KNN model is run using scikit-learn to predict the optimal delivery destination.

[0692] Step 5:

[0693] The server displays the suggested drop-off candidates on the display of the smart glasses. The input is the optimal drop-off candidate, and the output is the information to be displayed on the display of the smart glasses. The drop-off candidates are displayed on the display of the smart glasses in real time and notified to the user.

[0694] In this way, the sorting of mail and parcels with unknown destinations can be made more efficient, the risk of misdelivery can be reduced, and the accuracy of the sorting process can be improved.

[0695] Example 2

[0696] Next, a description will be given of Example 2 of the embodiment 2. In the following description, the data processing device 12 is referred to as a "server" and the smart glasses 214 are referred to as a "terminal".

[0697] Sorting mail and parcels with unknown addresses often depends on the work experience and sorting lists of the person in charge, which can be inefficient. In addition, there is a risk that the mail will be delivered to the wrong address because the details of past mailing destinations and deliveries cannot be properly utilized. Furthermore, there is an issue that it is difficult to suggest possible mailing destinations based on the sender and the type, size, and format of the delivery.

[0698] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0699] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for generating possible destinations using a generative AI model. This makes it possible to efficiently sort mail and parcels with unknown addresses and prevent delivery to the wrong destination.

[0700] "Sender" refers to the person or organization sending the mail or package.

[0701] "Delivery" refers to goods or documents sent by mail or courier.

[0702] "Type" indicates the classification of the delivery item, and refers to categories such as documents, parts, and goods.

[0703] "Size" refers to the size of the delivery item, such as small, medium, large, etc.

[0704] "Form" refers to the shape of the delivery item and the state of packaging, such as a box, envelope, or pallet.

[0705] "Potential drop-off locations" refer to locations or departments where a delivery may be dropped off.

[0706] "Previous delivery destination" refers to a location or department where a delivery has previously been delivered.

[0707] "Detailed information" refers to specific data and historical information about the delivery.

[0708] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to analyze data and generate appropriate recipient candidates.

[0709] "Database" refers to a system for storing and managing past mailing history and detailed delivery information.

[0710] "User" refers to the person or organization that operates the system to input data and review / decide on suggested recipients.

[0711] The present invention is a system for streamlining the sorting of mail and parcels with unknown addresses. This system is equipped with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. A specific embodiment of this system will be described below.

[0712] Hardware and Software Used

[0713] Hardware: Servers, terminals (PCs, tablets, etc.)

[0714] Software: Generative AI models (e.g. GPT-4), database management systems (e.g. MySQL), postal and parcel management systems

[0715] System program processing

[0716] 1. Data Entry

[0717] The user uses the terminal to enter data such as sender information and the type, size and format of the delivery item.

[0718] Example: A user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal and clicks the send button.

[0719] 2. Database Reference

[0720] The server uses a database management system (e.g. MySQL) to look up details of past mailings and deliveries.

[0721] Example: The server executes an SQL query and sends the following command to the database: "SELECT FROM mailing history WHERE sender = 'Company A' AND delivery = 'Parts of type B'".

[0722] 3. Generating recipient candidates

[0723] The server uses a generative AI model (e.g. GPT-4) to generate potential recipients based on the input data and past data.

[0724] Example: The server sends a prompt to the generative AI model: "If the sender is Company A and the delivery is a type B part, please suggest an appropriate delivery destination based on past data." The generative AI model responds: "Department C is appropriate."

[0725] 4. Suggestions for possible recipients

[0726] The server transmits the generated candidate mailbox destinations to the terminal and suggests them to the user.

[0727] Example: The server sends a message to the terminal saying, "We suggest posting to Department C." The content of the suggestion is displayed on the terminal screen.

[0728] 5. User confirmation and final decision

[0729] The user checks the proposed destination candidates and decides on the final destination.

[0730] Example: The user checks the proposal displayed on the terminal screen and clicks the "Approve posting to Department C" button. The user's selection is sent to the server, and the final posting destination is confirmed.

[0731] Examples of prompt statements

[0732] If the sender is "Company A", the delivery item is "Type B Parts" and the size is "Medium", please suggest an appropriate drop-off location based on past data.

[0733] In this way, users can efficiently sort mail and parcels that are not addressed to their destination. By utilizing generative AI models and a database management system, the system can accurately suggest possible recipients based on past data.

[0734] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0735] Step 1:

[0736] Data Entry

[0737] The user uses the terminal to enter data such as sender information and the type, size and format of the delivery item.

[0738] Input: sender information (e.g. Company A), type of delivery (e.g. Type B parts), size (e.g. medium size)

[0739] Specific operation: The user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal and clicks the send button.

[0740] Output: The entered data is sent to the server.

[0741] Step 2:

[0742] Database Reference

[0743] The server uses a database management system (e.g. MySQL) to look up details of past mailings and deliveries.

[0744] Input: Data entered by the user (sender information, type of delivery, size)

[0745] Specific operation: The server executes an SQL query and sends the command "SELECT FROM mailing history WHERE sender = 'Company A' AND delivery = 'Parts of type B'" to the database.

[0746] Output: Detailed information about past mailing destinations and deliveries is obtained.

[0747] Step 3:

[0748] Generating recipient candidates

[0749] The server uses a generative AI model (e.g. GPT-4) to generate potential recipients based on the input data and past data.

[0750] Input: Data entered by the user, details of past mailings and deliveries retrieved from a database

[0751] Specific operation: The server sends a prompt to the generative AI model: "If the sender is Company A and the delivery is a type B part, please suggest an appropriate delivery destination based on past data." The generative AI model responds: "Department C is appropriate."

[0752] Output: A list of possible recipients (e.g. Department C) is generated.

[0753] Step 4:

[0754] Suggestions for possible recipients

[0755] The server transmits the generated candidate mailbox destinations to the terminal and suggests them to the user.

[0756] Input: Generated recipient candidates (e.g. Department C)

[0757] Specific operation: The server sends a message to the terminal saying, "We suggest posting to Department C." The content of the suggestion is displayed on the terminal screen.

[0758] Output: The suggestions are displayed to the user.

[0759] Step 5:

[0760] User confirmation and final decision

[0761] The user checks the proposed destination candidates and decides on the final destination.

[0762] Input: Suggested recipient (e.g. Department C)

[0763] Specific operation: The user checks the proposal displayed on the terminal screen and clicks the "Approve submission to Department C" button. The user's selection is sent to the server.

[0764] Output: The final posting destination is confirmed.

[0765] In this way, users can efficiently sort mail and parcels that are not addressed to their destination. By utilizing generative AI models and a database management system, the system can accurately suggest possible recipients based on past data.

[0766] (Application example 2)

[0767] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server", and the smart glasses 214 will be referred to as a "terminal".

[0768] The conventional method of sorting mail and parcels with unknown addresses requires the worker to check the work experience and sorting list each time, which is inefficient and prone to human error. In addition, there is no system that uses past data to suggest the optimal destination, so there is a need to improve the accuracy and efficiency of the sorting process.

[0769] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0770] In this invention, the server includes a means for suggesting possible post destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible post destinations based on past post destinations and detailed information on the delivery, a means installed in the smartphone for suggesting the optimal post destination based on past data, and a means for saving the history of the suggested post destinations and improving the accuracy of the suggestions from the next time onwards. This makes it possible to significantly improve the efficiency and accuracy of the sorting work for mail and parcels with unknown addresses.

[0771] "Unknown address mail or parcel" refers to mail or parcel that cannot be delivered to a specific address through normal delivery procedures because the address specified by the sender is unclear.

[0772] "Sorting work" refers to the task of classifying mail and parcels based on their destination and content, and then distributing them to the appropriate destination.

[0773] "Sender" refers to the individual or organization sending the mail or parcel.

[0774] "Type of delivery" refers to the category of the contents of a mail or parcel, such as documents, merchandise, parts, etc.

[0775] "Size" refers to the physical size of a mail item or parcel, and is classified as small, medium, large, etc.

[0776] "Format" refers to the shape and type of packaging of a mail or parcel, such as an envelope, box, or pallet.

[0777] "Potential drop-off locations" refer to locations or departments to which undeliverable mail or parcels may be delivered.

[0778] "Previous mailing destination" refers to a location or department to which undeliverable mail or parcels have previously been delivered.

[0779] "Detailed information" refers to specific data and history about mail or parcels.

[0780] A "smartphone" is a type of mobile phone and a multi-function device that can connect to the Internet and install applications.

[0781] "Saving history" means recording information about past mailing destinations and deliveries in a database so that it can be referenced later.

[0782] "Suggestion accuracy" refers to the system's ability to accurately suggest the optimal drop-off destination.

[0783] As an embodiment of the present invention, a system for streamlining the sorting of mail and parcels with unknown destinations is constructed. A specific embodiment of the system is described below.

[0784] System Configuration

[0785] Hardware:

[0786] Smartphone (iOS or Android)

[0787] software:

[0788] Python

[0789] Pandas (Data Processing)

[0790] Scikit-learn (machine learning model)

[0791] Program processing explanation

[0792] Data Loading:

[0793] The server reads past delivery data from a CSV file, which gives information about the sender, type of delivery, size, format, and past delivery destination.

[0794] Split features and labels:

[0795] Based on the data it reads, the server classifies the sender, type of delivery, size, and format as features, and the delivery destination as a label.

[0796] Model training:

[0797] The server trains a machine learning model using K-nearest neighbors (KNN) to improve its ability to suggest optimal drop destinations based on past data.

[0798] Suggested destinations:

[0799] When a user uses their smartphone to input information about a new delivery (sender, type of delivery, size, format), the server uses the trained model to suggest the optimal delivery location.

[0800] Save history:

[0801] The server stores the history of suggested destinations in a database, which improves the accuracy of suggestions from the next time onwards.

[0802] Examples

[0803] For example, if the sender is "Company A," the delivery item is a "Type B part," the size is "medium," and the format is "box," it will suggest posting it to "Department C" based on past data.

[0804] Examples of prompt statements

[0805] Please use the following information to suggest the best place to send your message.

[0806] From: Company A

[0807] Delivery type: B-type parts

[0808] Size: Medium

[0809] Format: Box

[0810] In this way, a system can be realized that improves the efficiency and accuracy of sorting mail and parcels with unknown destinations.

[0811] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0812] Step 1:

[0813] Data Import

[0814] The server reads past delivery data from a CSV file. The path of the CSV file is specified as input. The server reads the CSV file in data frame format using the Pandas library and obtains information on the sender, type of delivery, size, format, and past recipients. A data frame is generated as output.

[0815] Step 2:

[0816] Splitting features and labels

[0817] Based on the data read, the server separates the sender, type, size, and format of the delivery as features, and the recipient as a label. The data frame generated in step 1 is used as input. The server splits the features and labels into separate data frames, in preparation for training the machine learning model. As output, a feature data frame and a label data frame are generated.

[0818] Step 3:

[0819] Model Training

[0820] The server trains a machine learning model using K-nearest neighbors (KNN). As input, the feature and label dataframes generated in step 2 are used. The server trains the KNN model using the Scikit-learn library to improve its ability to suggest optimal drop destinations based on past data. As output, a trained KNN model is generated.

[0821] Step 4:

[0822] Suggested destination

[0823] A user inputs new parcel information (sender, parcel type, size, format) using a smartphone. The parcel information entered by the user is used as input. The server uses the trained KNN model to suggest the optimal delivery destination based on the input information. As output, the suggested delivery destination is displayed to the user.

[0824] Step 5:

[0825] History storage

[0826] The server saves the history of suggested destinations in a database. As input, the proposed destinations in step 4 and the delivery information entered by the user are used. The server records this information in the database to improve the accuracy of future suggestions. As output, the history data is saved in the database.

[0827] In this way, a system can be realized that improves the efficiency and accuracy of sorting mail and parcels with unknown destinations.

[0828] In addition, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0829] "Example 1"

[0830] As an embodiment of the present invention, a system incorporating an emotion engine is provided. The system recognizes the user's emotion and adjusts the candidate destinations based on the emotion. Specifically, if the system recognizes that the user is stressed, the system reduces the number of candidate destinations and simplifies the options. Also, if the system recognizes that the user is relaxed, the system provides more candidate destinations, giving the user more freedom of choice. Furthermore, the emotion engine also adjusts the way in which the candidate destinations are presented based on the user's emotion. For example, if the system recognizes that the user is anxious, the system does not display a list of candidate destinations, but directly presents the most likely destination. This allows the user to make a decision quickly.

[0831] "Example 2"

[0832] As an embodiment of the present invention, a system incorporating an emotion engine is provided. The system recognizes the user's emotion and adjusts the candidate destinations based on the emotion. Specifically, if the system recognizes that the user is stressed, the system reduces the number of candidate destinations and simplifies the options. Also, if the system recognizes that the user is relaxed, the system provides more candidate destinations, giving the user more freedom of choice. Furthermore, the emotion engine also adjusts the way in which the candidate destinations are presented based on the user's emotion. For example, if the system recognizes that the user is anxious, the system does not display a list of candidate destinations, but directly presents the most likely destination. This allows the user to make a decision quickly.

[0833] The process flow of each embodiment will be described below.

[0834] "Example 1"

[0835] Step 1: The system recognizes the user's emotions using an emotion engine, which infers emotions from the user's facial expressions, tone of voice, choice of words, etc.

[0836] Step 2: After the emotion engine recognizes the user's emotion, the system will adjust the suggested destinations based on the emotion. For example, if the system recognizes that the user is stressed, it will reduce the number of suggested destinations and simplify the options.

[0837] Step 3: On the other hand, if the system detects that the user is relaxed, it will provide more possible destinations, giving the user more freedom of choice.

[0838] Step 4: Furthermore, the emotion engine will also adjust the way in which the system presents the possible destinations based on the user's emotions. For example, if the system detects that the user is in a hurry, it will directly present the most likely destination instead of displaying a list of possible destinations, allowing the user to make a decision quickly.

[0839] Example 1

[0840] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal".

[0841] Conventional methods for sorting mail and parcels with unknown addresses are inefficient because they require the person in charge to check the work experience and sorting list each time. In addition, there is a problem that appropriate support cannot be provided to users who feel stressed or impatient because the system cannot respond flexibly to the user's emotional state. This has led to a demand for more efficient sorting work and improved user satisfaction.

[0842] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0843] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on the emotions. This makes it possible to streamline the sorting of mail and parcels with unknown addresses and to respond flexibly to the user's emotional state.

[0844] "Sender" refers to the individual or organization sending the mail or parcel.

[0845] "Delivery" refers to goods or documents sent by mail or courier.

[0846] "Type" is an attribute that indicates the classification of the delivery, and includes, for example, documents, parts, and merchandise.

[0847] "Size" is an attribute that indicates the physical size of a delivery item, and includes, for example, length, width, height, weight, and the like.

[0848] "Form" is an attribute that indicates the shape of the delivery item or the state of packaging, and includes, for example, a box, an envelope, a pallet, and the like.

[0849] "Potential drop-off locations" refers to locations or departments where deliveries may be dropped off.

[0850] "Previous mailbox locations" refers to records of locations or departments where deliveries have been previously mailed.

[0851] "Detailed information" refers to specific data and history about the delivery.

[0852] "Emotion recognition" refers to identifying a user's emotional state by analyzing their facial expressions and tone of voice.

[0853] "Adjust" refers to changing the system's behavior or suggestions to suit specific conditions or circumstances.

[0854] The present invention is a system for streamlining the sorting of mail and parcels with unknown destinations. This system includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on those emotions.

[0855] Hardware and Software Used

[0856] Hardware

[0857] Server: Analyzes and stores data.

[0858] Terminal: Provides a user interface and is equipped with a camera and microphone for recognizing the user's emotions.

[0859] Scanner: Scan postal and courier labels.

[0860] software

[0861] Google Cloud Vision API: Converts paper data into digital data using OCR (optical character recognition) technology.

[0862] MySQL: A database management system that stores analysis results and past submission data.

[0863] Microsoft Azure Emotion API: Recognizes emotions by analyzing the user's facial expressions and tone of voice.

[0864] System Operation

[0865] Data Entry and Analysis

[0866] The server receives data such as sender information for mail and parcels, and the type, size, and format of the delivery. Specifically, it scans the mail and parcel labels and converts them from paper to digital data using Google Cloud Vision API. The server analyzes the received data and identifies the sender and the type, size, and format of the delivery. The analysis results are stored in a MySQL database.

[0867] Suggestions for possible recipients

[0868] The server searches a database of past delivery destinations based on the analysis results and suggests the most appropriate delivery destination. For example, if the sender is "Company A" and the delivery item is "Type B parts," it suggests delivery to "Department C" based on past data. Also, if the size of the delivery item falls within a certain range, it suggests delivery to a specific location.

[0869] Emotion engine adjustment

[0870] The device uses a camera and microphone to analyze the user's facial expressions and tone of voice to recognize the user's emotions. It uses Microsoft Azure's Emotion API. If the server recognizes that the user is stressed, it reduces the number of possible delivery candidates and simplifies the options. Conversely, if the server recognizes that the user is relaxed, it provides more possible delivery candidates.

[0871] Suggestion of possible recipients

[0872] The server sends the adjusted drop-off candidates to the terminal and presents them to the user. If the server detects that the user is in a hurry, it will directly present the most likely drop-off destination, allowing the user to make a quick decision.

[0873] Examples of concrete examples and prompts

[0874] Examples

[0875] When a user receives a "Part B" from "Company A", the system will suggest dropping it off at "Division C" based on past data. If the system detects that the user is stressed, it will only suggest dropping it off at "Division C".

[0876] Examples of prompt statements

[0877] "We have received part B from company A. Please suggest the best delivery destination based on past data. Also, please adjust the suggestions based on user sentiment."

[0878] The above is an embodiment of the present invention.

[0879] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0880] Step 1:

[0881] Data Entry

[0882] The server receives data such as sender information for mail or parcels, and the type, size, and format of the delivery. As input, it scans the label of the mail or parcel to obtain image data. In concrete terms, the server obtains image data from the scanner and calls the Google Cloud Vision API to extract text information from the image data. As output, it obtains the extracted text information.

[0883] Step 2:

[0884] 4. Data Analysis

[0885] The server analyzes the text information obtained in step 1 to identify the sender, type of delivery, size, and format. As input, it receives the extracted text information. In concrete terms, the server analyzes the text information to identify the sender, type of delivery, size, and format. As output, it obtains the analysis results, which are stored in a MySQL database.

[0886] Step 3:

[0887] Suggestions for possible recipients

[0888] The server searches the database of past destinations based on the analysis results obtained in step 2, and proposes the most suitable destination candidates. It receives the analysis results as input. In concrete terms, the server queries the MySQL database to retrieve the past destination data. As output, it obtains a list of optimal destination candidates.

[0889] Step 4:

[0890] Emotion engine adjustment

[0891] To recognize the user's emotions, the device uses a camera and microphone to analyze the user's facial expressions and tone of voice. It receives the user's facial expression and voice data as input. Specifically, the device calls Microsoft Azure's Emotion API to analyze the user's emotions. The output is the user's emotional state. The server adjusts the number of possible recipients based on the user's emotional state.

[0892] Step 5:

[0893] Suggestion of possible recipients

[0894] The server sends the adjusted list of destination candidates in step 4 to the terminal and presents them to the user. As input, it receives the adjusted list of destination candidates. In concrete terms, the server sends the adjusted destination candidates to the terminal, and the terminal displays the destination candidates on the display. As output, the destination candidates presented to the user are obtained.

[0895] (Application example 1)

[0896] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal".

[0897] The conventional method for sorting mail and parcels with unknown addresses requires the person in charge to check the work experience and sorting list each time, which makes it inefficient and prone to errors. In addition, there are many situations that require flexible responses based on the user's emotions, and there was no system that could handle this.

[0898] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0899] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the number of possible destinations and the presentation method based on the emotions. This makes it possible to streamline the work of sorting mail and parcels with unknown addresses and to respond flexibly to the user's emotions.

[0900] "Sender" refers to the person or organization sending the mail or package.

[0901] "Delivery" refers to goods or documents sent by mail or courier.

[0902] "Type" is an attribute that indicates the classification of the delivery, such as documents, parts, or merchandise.

[0903] "Size" is an attribute that indicates the size or dimensions of the delivery item.

[0904] "Form" is an attribute that indicates the shape of the delivery and the state of packaging.

[0905] "Possible drop-off locations" refers to possible locations where deliveries may be dropped off or potential recipients.

[0906] "Previous delivery destination" refers to a location or recipient where a delivery has previously been delivered.

[0907] "Detailed information" refers to specific data and historical information about the delivery.

[0908] "Emotion recognition" refers to detecting and understanding a user's emotional state.

[0909] "Emotion-based adjustment" refers to changing the system's behavior or presentation in response to recognized emotions.

[0910] A system for implementing the present invention includes a server, a smartphone, an emotion recognition engine, and a machine learning model. Specific embodiments are described below.

[0911] System Configuration

[0912] Hardware

[0913] Server: Stores and processes data.

[0914] Smartphones: Used as a user interface and receive user input.

[0915] Emotion recognition engine: Recognize user emotions in real time.

[0916] Machine learning model: A model for suggesting possible recipients (e.g., KNN model).

[0917] software

[0918] Python: Used to implement the program.

[0919] scikit-learn: Used to train and predict machine learning models.

[0920] emotion_recognition: A library of emotion recognition engines.

[0921] Data processing and calculation

[0922] server

[0923] The server stores past delivery destination data and delivery details, retrieves them when needed, and also trains machine learning models to suggest delivery destinations based on user input.

[0924] Smartphone

[0925] The smartphone receives input from the user (sender, type, size, and format of the delivery item) and sends it to the server. It also recognizes the user's emotions in real time through an emotion recognition engine and sends that information to the server.

[0926] Emotion Recognition Engine

[0927] The emotion recognition engine recognizes the user's emotions from their facial expressions and voice, and returns the results to the smartphone, which then adjusts the number of possible recipients and the way they are presented according to the user's emotional state.

[0928] Machine learning models

[0929] A machine learning model (e.g., KNN model) is trained based on past data and suggests optimal delivery candidates based on user input data. It receives information from an emotion recognition engine and adjusts the suggestions according to the user's emotions.

[0930] Examples

[0931] For example, if the sender is "Company A," the delivery is "Type B parts," and the size is "medium," the system will suggest dropping it off at "Department C" based on past data. If the user is feeling stressed, the system will narrow down the suggestions to one, and if the user is relaxed, it will display up to five suggestions.

[0932] Examples of prompt statements

[0933] Suggest possible delivery destinations for a "medium" size package with a "Type B part" from "Company A". If the user is stressed, narrow down the suggestions to one, and if the user is relaxed, display up to five suggestions.

[0934] In this way, a system is provided that can streamline the sorting of mail and parcels with unknown destinations and can flexibly respond to the user's emotions.

[0935] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0936] Step 1:

[0937] The user uses a smartphone to input the sender, type of delivery, size, and format. The input data is sent from the smartphone to the server. The input data includes the sender name, type of delivery, size, and format.

[0938] Step 2:

[0939] The smartphone uses an emotion recognition engine to recognize the user's emotions in real time. The emotion recognition engine analyzes the user's facial expressions and voice data to detect their emotional state (e.g., stress, relaxation). The detected emotion data is sent from the smartphone to a server.

[0940] Step 3:

[0941] The server retrieves relevant data from a database of past mailing destinations based on the received input data (sender, type of delivery, size, format) and emotion data. The server compares the past mailing destination data with the detailed delivery information to generate optimal mailing destination candidates.

[0942] Step 4:

[0943] The server uses a machine learning model (e.g., KNN model) to suggest possible destinations based on the input data. The machine learning model is trained based on past data and analyzes the input data to predict the best possible destinations. The predicted destination candidates are sent from the server to the smartphone.

[0944] Step 5:

[0945] The server adjusts the number and presentation of possible recipients based on the emotion data. For example, if the user is stressed, the server reduces the number of possible recipients and presents only one most likely recipient. If the user is relaxed, the server presents multiple possible recipients (e.g., five).

[0946] Step 6:

[0947] The smartphone displays the candidate destinations received from the server to the user. The user selects the most suitable destination from the presented candidates. The selected destination is sent from the smartphone to the server.

[0948] Step 7:

[0949] The server records the destination selected by the user and updates the database to improve the accuracy of the suggestions for the next time, allowing the system to continuously learn and improve the accuracy of the suggestions.

[0950] In this way, a system is provided that can streamline the sorting of mail and parcels with unknown destinations and can flexibly respond to the user's emotions.

[0951] Example 2

[0952] Next, a description will be given of Example 2 of the embodiment 2. In the following description, the data processing device 12 is referred to as a "server" and the smart glasses 214 are referred to as a "terminal".

[0953] Conventional methods for sorting mail and parcels with unknown addresses rely on the work experience and sorting lists of the person in charge, which makes them inefficient and prone to errors. In addition, there is also the problem that the system is unable to respond flexibly to the user's emotional state, making it difficult to provide appropriate support to users who feel stressed or impatient. Furthermore, there are insufficient suggestions for delivery destinations based on the sender and the type, size, and format of the delivery, making it difficult to quickly find the optimal delivery destination.

[0954] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0955] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on the emotions. This makes it possible to streamline the sorting of mail and parcels with unknown destinations and reduce errors. In addition, since the server can flexibly respond to the user's emotional state, it reduces the user's stress and impatience and allows the user to quickly find the optimal destination.

[0956] "Sender" refers to the person or organization sending the mail or package.

[0957] "Delivery" refers to goods or documents sent by mail or courier.

[0958] "Type" refers to the classification or category of the delivery, such as documents, parts, merchandise, etc.

[0959] "Size" refers to the size or dimensions of the delivery item, and includes, for example, small, medium, large, etc.

[0960] "Format" refers to the shape or packaging of the delivery, such as a box, envelope, or pallet.

[0961] "Potential delivery destinations" refers to potential locations or departments where deliveries may be dropped off.

[0962] "Previous delivery destination" refers to a location or department where a delivery has previously been delivered.

[0963] "Detailed information" refers to specific data and historical information about the delivery.

[0964] "Emotion recognition engine" refers to software or hardware for analyzing and recognizing a user's emotional state.

[0965] A "generative AI model" refers to a model that uses artificial intelligence to analyze data and generate optimal results.

[0966] A "prompt" refers to an instruction or question that is input into a generative AI model.

[0967] The present invention is a system for streamlining the sorting of mail and parcels with unknown addresses, and includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery item, a means for suggesting possible destinations based on past delivery destinations and detailed information about the delivery item, and a means for recognizing the user's emotions and adjusting the possible destinations based on those emotions.

[0968] Hardware and Software Used

[0969] Hardware: Servers, user devices (PCs, tablets, smartphones)

[0970] Software: Generative AI models, emotion recognition engine, database management system (DBMS)

[0971] Specific operation of the system

[0972] 1. Data Entry:

[0973] The user uses the terminal to enter information such as the sender and the type, size, and format of the delivery.

[0974] Specific operation: The user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal.

[0975] 2. Data Analysis:

[0976] The server receives the entered data and compares it with its historical database.

[0977] Specific operation: The server uses a database management system (DBMS) to search for data that "Parts of type B from Company A were previously mailed to Department C."

[0978] 3. Suggestions for possible recipients:

[0979] The server uses a generative AI model to suggest optimal delivery candidates.

[0980] Specific operation: The server provides input data as prompts to the generative AI model to generate optimal delivery candidates.

[0981] Example: The server inputs the prompt "Sender: Company A, Delivery: Type B part, Size: Medium" into the generative AI model and outputs "Department C" as a possible recipient.

[0982] 4. Emotion recognition:

[0983] The terminal uses an emotion recognition engine to analyze the user's emotions in real time while the user is operating the terminal.

[0984] Specific behavior: Uses the device's camera and microphone to analyze the user's facial expressions and tone of voice to determine their emotional state.

[0985] Example: The device captures the user's face with a camera and an emotion recognition engine detects "stress."

[0986] 5. Adjustment of possible recipients:

[0987] The server adjusts the number of possible recipients and the presentation method based on the user's feelings.

[0988] What it does: The server receives data from the emotion recognition engine and refines the list of potential recipients.

[0989] Example: The server receives information that "the user is feeling stressed" and narrows down the possible recipients to only "Department C".

[0990] Examples

[0991] Example of input prompt:

[0992] "If the sender is Company A and the delivery item is Type B parts, please suggest the best delivery location based on past data."

[0993] "If users are feeling stressed, present them with fewer possible recipients."

[0994] In this way, the system suggests optimal candidates for recipients based on the user's input data and emotional state.

[0995] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0996] Step 1:

[0997] Data Entry

[0998] The user uses the terminal to enter information such as the sender and the type, size, and format of the delivery.

[0999] Input: sender, delivery type, size, format

[1000] Specific operation: The user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal.

[1001] Output: The entered data is sent to the server.

[1002] Step 2:

[1003] 4. Data Analysis

[1004] The server receives the entered data and compares it with its historical database.

[1005] Input: Data sent by the user, such as sender, type of delivery, size, and format

[1006] Specific operation: The server uses a database management system (DBMS) to search for data that "Parts of type B from Company A were previously mailed to Department C."

[1007] Output: The matching result is the past mailing destination data.

[1008] Step 3:

[1009] Suggestions for possible recipients

[1010] The server uses a generative AI model to suggest optimal delivery candidates.

[1011] Input: Past mailing destination data obtained as a result of matching, and data input by users

[1012] Specific operation: The server provides input data as prompts to the generative AI model to generate optimal delivery candidates.

[1013] Output: A list of potential recipients from the generative AI model.

[1014] Step 4:

[1015] emotion recognition

[1016] The terminal uses an emotion recognition engine to analyze the user's emotions in real time while the user is operating the terminal.

[1017] Input: User facial expressions and tone of voice

[1018] Specific behavior: Uses the device's camera and microphone to analyze the user's facial expressions and tone of voice to determine their emotional state.

[1019] Output: Emotion state data from the emotion recognition engine.

[1020] Step 5:

[1021] Adjustment of possible recipients

[1022] The server adjusts the number of possible recipients and the presentation method based on the user's feelings.

[1023] Input: Emotion state data from an emotion recognition engine, a list of recipient candidates from a generative AI model

[1024] What it does: The server receives data from the emotion recognition engine and refines the list of potential recipients.

[1025] Output: The refined list of possible recipients is presented to the user.

[1026] In this way, the system suggests optimal candidates for recipients based on the user's input data and emotional state.

[1027] (Application example 2)

[1028] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server", and the smart glasses 214 will be referred to as a "terminal".

[1029] Sorting mail and parcels with unknown addresses is difficult because it is difficult to find the appropriate place to drop them off based on the sender and the type, size, and format of the parcel, and because it requires the staff member to check the work experience and sorting list each time.Furthermore, flexible responses according to the user's emotional state are required, but conventional systems were unable to achieve this.

[1030] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1031] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information of the delivery, a means for recognizing the user's emotions and adjusting the possible destinations based on the emotions, a means for scanning the delivery with a camera in the smart glasses and recognizing the sender and the type, size, and format of the delivery, and a means for presenting the optimal possible destinations based on the recognized information, past data, and the user's emotions. This makes it possible to efficiently and flexibly sort mail and parcels with unknown addresses.

[1032] A "sender" is a person or organization that sends a letter or package.

[1033] A "delivery" is an item sent by mail or courier.

[1034] "Type" is an attribute that indicates the classification of the delivery.

[1035] "Size" is an attribute that indicates the size of the delivery item.

[1036] "Form" is an attribute that indicates the shape of the delivery item and the type of packaging.

[1037] "Potential delivery destinations" are potential locations or departments where deliveries may be delivered.

[1038] A "previous delivery destination" is a location or department where a delivery has been previously delivered.

[1039] "Detailed information" refers to specific data and history information regarding the delivery.

[1040] "User emotion" refers to the psychological state of a person using a system.

[1041] "Smart glasses" are glasses-type devices that have built-in cameras and displays and have the ability to display information.

[1042] A "camera" is a device that captures images or videos.

[1043] "Recognition" refers to the system analyzing input information and determining specific attributes or conditions.

[1044] "Data" is a collection of information that is processed by a system.

[1045] "Presentation" means that the system displays information to the user.

[1046] A system for implementing the present invention includes smart glasses, a camera, an emotion recognition engine, a database, and a display for presenting possible recipients. The specific configuration and operation of the system will be described below.

[1047] System Configuration

[1048] 1. Smart glasses: The smart glasses are equipped with a camera and a display and are worn by workers. The camera scans the package and the display shows possible recipients.

[1049] 2. Camera: The camera built into the smart glasses takes pictures of the package and provides data to recognize the sender, package type, size and format.

[1050] 3. Emotion Recognition Engine: The emotion recognition engine scans the user's face and recognizes their emotions, allowing it to tailor the recipient suggestions to suit the user's psychological state.

[1051] 4. Database: The database stores detailed information on past mailing destinations and deliveries, and uses this data to suggest possible mailing destinations.

[1052] 5. A means for presenting possible recipients: A means for presenting optimal recipient candidates based on recognized information, past data, and user sentiment.

[1053] System Operation

[1054] 1. Parcel scanning: A worker wears smart glasses and scans the parcel with a camera, which captures an image of the parcel and identifies the sender, type, size and format of the parcel.

[1055] 2. Database query: Based on the recognized information, detailed information on past mailing destinations and deliveries is obtained from the database.

[1056] 3. Emotion Recognition: The camera in the smart glasses scans the user's face and the emotion recognition engine recognizes the user's emotions.

[1057] 4. Suggestion of possible recipients: Suggest optimal recipients based on recognized information, past data, and user emotions. If the user is stressed, the number of possible recipients will be reduced, and if the user is relaxed, more possible recipients will be provided.

[1058] 5. Presentation of possible drop-off destinations: Potential drop-off destinations are displayed on the smart glasses display, allowing the user to make a quick decision.

[1059] Examples

[1060] When a logistics center worker wears smart glasses and scans an unknown delivery, the system automatically recognizes the sender and the type, size, and format of the delivery, and suggests the best possible delivery destination based on past data and the user's emotions. For example, if the sender is "Company A" and the delivery is "Type B parts," the system will suggest delivery to "Department C" based on past data.

[1061] Examples of prompt statements

[1062] "If the sender is Company A and the delivery is for Type B parts, suggest the best possible delivery destinations based on past data. Also, if the user is stressed, provide fewer delivery destinations, and if the user is relaxed, provide more delivery destinations."

[1063] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1064] Step 1:

[1065] The user puts on the smart glasses and scans the delivery with the camera.

[1066] Input: Image of delivery

[1067] Output: Image data

[1068] Specific operation: The camera in the smart glasses takes a picture of the delivery item and captures the image data.

[1069] Step 2:

[1070] The server analyzes the image data and identifies the sender and the type, size and format of the delivery.

[1071] Input: Image data

[1072] Output: sender, delivery type, size, format information

[1073] What it does: Uses image recognition algorithms to extract sender, package type, size and format from image data.

[1074] Step 3:

[1075] Based on the information recognized by the server, detailed information about past mailing destinations and deliveries is retrieved from the database.

[1076] Input: sender, delivery type, size, and format information

[1077] Output: Details of past mailing destinations and deliveries

[1078] What it does: Executes database queries to retrieve details about past mailings and deliveries.

[1079] Step 4:

[1080] The user scans their face with the smart glasses' camera, and the emotion recognition engine recognizes the user's emotions.

[1081] Input: User's face image

[1082] Output: The user's emotional state.

[1083] What it does: It uses emotion recognition algorithms to analyze the user's emotional state from facial images.

[1084] Step 5:

[1085] The server suggests optimal candidates for delivery destinations based on the information recognized by the server, past data, and user sentiment.

[1086] Input: sender, delivery type, size, format information, past mailing destinations and delivery details, user emotional state

[1087] Output: List of possible recipients

[1088] What it does: It uses a generative AI model to analyze input data and generate optimal delivery candidates.

[1089] Step 6:

[1090] The server displays possible delivery destinations on the smart glasses' display.

[1091] Input: List of possible recipients

[1092] Output: Possible recipients displayed on the smart glasses display

[1093] Specific operation: The list of possible recipients is sent to the smart glasses display and displayed to the user.

[1094] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1095] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1096] Another example of generative AI is Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1097] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1098] [Third embodiment]

[1099] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1100] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1101] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a wide area network (WAN) and / or a local area network (LAN).

[1102] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1103] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[1104] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[1105] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[1106] Fig. 6 shows an example of main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1107] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1108] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1109] In the headset type terminal 314, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1110] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1111] "Example 1"

[1112] The system of the present invention is for sorting mail and parcels with unknown addresses. Specifically, it is provided with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. For example, if the sender is "Company A" and the delivery is "Parts Type B," it is suggested to post the delivery to "Department C" based on past data. In addition, if the size of the delivery falls within a specific range, it is possible to suggest possible destinations according to various conditions, such as suggesting posting to a specific location.

[1113] "Example 2"

[1114] The system of the present invention is for sorting mail and parcels with unknown addresses. Specifically, it is provided with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. For example, if the sender is "Company A" and the delivery is "Parts Type B," it is suggested to post the delivery to "Department C" based on past data. In addition, if the size of the delivery falls within a specific range, it is possible to suggest possible destinations according to various conditions, such as suggesting posting to a specific location.

[1115] The process flow of each embodiment will be described below.

[1116] "Example 1"

[1117] Step 1: The system receives information about the bounced mail or parcel, including the sender, type, size, and format of the parcel.

[1118] Step 2: Next, the system will suggest possible delivery destinations based on the received information. Specifically, it will suggest possible delivery destinations based on the sender and the type, size and format of the delivery.

[1119] Step 3: The system then suggests possible delivery destinations based on past delivery destinations and detailed information about the delivery item. For example, if the sender is "Company A" and the delivery item is "Parts Type B," the system will suggest delivery to "Department C" based on past data.

[1120] Step 4: Finally, the system presents the proposed destination candidates to the user, who can select the most suitable destination from the presented destination candidates.

[1121] Example 1

[1122] Next, a description will be given of Example 1 of the embodiment 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal".

[1123] Sorting mail and parcels with unknown addresses requires the staff to check their work experience and sorting list each time, which is inefficient and increases the risk of misdelivery. There is also the problem that it is difficult to quickly find the appropriate delivery destination based on the sender and the type, size, and format of the delivery. To solve these issues, a system is needed that utilizes past data to efficiently and accurately suggest delivery destination candidates.

[1124] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1125] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, a means for generating possible destinations using a generative AI model, a means for transmitting data entered by a user to the server, a means for the server to acquire past data from a database, a means for the server to transmit the proposed results to a terminal, and a means for the user to check the proposed results and select an appropriate destination. This makes it possible to efficiently and accurately sort mail and parcels with unknown addresses.

[1126] "Sender" refers to the person or organization sending the mail or package.

[1127] "Delivery" refers to items sent by mail or courier service.

[1128] "Type" is an attribute that indicates the classification of the delivery.

[1129] "Size" is an attribute that indicates the size of the delivery item.

[1130] "Form" is an attribute that indicates the shape of the delivery item and the type of packaging.

[1131] "Possible delivery destinations" refer to locations or departments where deliveries may be dropped off.

[1132] "Previous mailbox location" refers to a location or department where a delivery has been previously mailed.

[1133] "Detailed information" refers to specific data and history about the delivery.

[1134] A "generative AI model" is a model that uses artificial intelligence to analyze data and generate potential recipients.

[1135] A "user" is a person who operates the system to enter postal or courier information.

[1136] A "server" is a computer system that processes and stores data.

[1137] A "terminal" is a computer that includes input and display devices operated by a user.

[1138] A "database" is a system for storing detailed information about past mailing destinations and deliveries.

[1139] "Suggested results" refer to the candidate recipients generated by the generative AI model.

[1140] This invention is a system for efficiently and accurately sorting mail and parcels with unknown destinations. This system is realized using a server, a terminal, a generative AI model, and a database management system (DBMS).

[1141] Hardware and Software Used

[1142] Hardware: Servers, terminals

[1143] Software: Generative AI models, Database Management System (DBMS)

[1144] System Configuration

[1145] 1. User enters postal or courier information:

[1146] The user inputs information about the mail or parcel they received, such as the sender, type, size, and format of the parcel, into the input form on the terminal. For example, the sender is "Company A," the parcel is "Type B parts," the size is "medium," and the format is "box."

[1147] 2. The device sends the input data to the server:

[1148] The terminal sends the data entered by the user to the server, specifically, using an HTTP request to send information such as the sender, type of delivery, size, format, etc.

[1149] 3. The server retrieves historical data from the database:

[1150] The server uses a database management system (DBMS) to retrieve past mailing destinations and delivery details. For example, it executes an SQL query to retrieve past mailing destination data for "Parts of type B" from "Company A."

[1151] 4. The server uses the generative AI model to generate possible recipients:

[1152] The server inputs the acquired past data into the generative AI model to generate candidates for recipients. Specifically, it inputs the following prompt sentence to the generative AI model:

[1153] If the sender is "Company A" and the delivery is "Parts Type B", which department would you suggest to post it to based on past data?

[1154] Based on this prompt, the generative AI model suggests sending the item to "Division C."

[1155] 5. The server sends the proposal results to the device:

[1156] The server sends the candidate mailing destinations generated by the generative AI model to the terminal. Specifically, it uses an HTTP response to send the result proposing mailing to "Department C" to the terminal.

[1157] 6. User reviews the suggestions and selects the appropriate destination:

[1158] The user checks the proposed results displayed on the terminal. For example, if the proposal is to send the item to "Division C," the user can accept the proposal or select another destination. The destination selected by the user is recorded in the system and used as reference data for the future.

[1159] In this way, users can efficiently find the correct drop-off location for undelivered mail or parcels, improving the efficiency of sorting operations and reducing the risk of misdelivery.

[1160] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1161] Step 1:

[1162] A user enters postal or courier information.

[1163] The user inputs information about the mail or parcel they received, such as the sender, type of item, size, and format, into the input form on the terminal. For example, the sender is "Company A," the item is "Type B parts," the size is "medium," and the format is "box." The input data is temporarily stored in the terminal's memory.

[1164] Step 2:

[1165] The terminal transmits the input data to the server.

[1166] The terminal sends the data entered by the user to the server using an HTTP request. Specifically, it sends JSON-formatted data including information such as the sender, type of delivery, size, and format to the server. The server receives and analyzes this data.

[1167] Step 3:

[1168] The server retrieves historical data from the database.

[1169] The server uses a database management system (DBMS) to retrieve past mailing destinations and detailed delivery information. For example, it executes an SQL query to retrieve past mailing destination data for "Part B" from "Company A." The retrieved data is stored in the server's memory.

[1170] Step 4:

[1171] The server uses the generative AI model to generate candidate recipients.

[1172] The server inputs the acquired past data into the generative AI model to generate candidates for recipients. Specifically, it inputs the following prompt sentence to the generative AI model:

[1173] If the sender is "Company A" and the delivery is "Parts Type B", which department would you suggest to post it to based on past data?

[1174] Based on this prompt, the generative AI model suggests sending the mail to "Division C." The generated mail destination candidates are stored in the server's memory.

[1175] Step 5:

[1176] The server transmits the proposal results to the terminal.

[1177] The server sends the candidate mailing destinations generated by the generative AI model to the terminal using an HTTP response. Specifically, it sends JSON-formatted data including the results of suggesting mailing to "Department C" to the terminal. The terminal receives and analyzes this data.

[1178] Step 6:

[1179] The user checks the suggestions and selects the appropriate destination.

[1180] The user checks the proposal results displayed on the terminal. For example, if the proposal is to send the item to "Division C," the user can accept the proposal or select another destination. The destination selected by the user is sent from the terminal to the server and recorded in the system. This will be used as reference data in the future.

[1181] (Application example 1)

[1182] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server", and the headset type terminal 314 will be referred to as a "terminal".

[1183] The conventional method of sorting mail and parcels with unknown addresses relies on the experience of the person in charge and on sorting lists, which results in low efficiency and a high risk of misdelivery. In addition, there is a problem of reduced accuracy in sorting work because there are insufficient suggestions for the delivery destination based on the sender and the type, size, and format of the delivery.

[1184] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means. In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery item, a means for suggesting possible destinations based on past destinations and detailed information of the delivery item, a means for scanning the delivery item using an image processing device and recognizing the sender and the type and size of the delivery item, and a means for suggesting optimal possible destinations in real time using a machine learning model. This makes it possible to streamline the sorting of mail and parcels with unknown destinations, reduce the risk of misdelivery, and improve the accuracy of the sorting work.

[1185] "Undelivered mail or parcel" refers to mail or parcel whose delivery destination cannot be identified using normal delivery procedures because the recipient's address or name is unclear.

[1186] "Sorting" refers to the task of classifying mail and parcels according to their destination and content, and then distributing them to the appropriate destination.

[1187] A "sender" is an individual or organization that sends mail or parcels.

[1188] "Type of delivery" refers to the classification of the contents of a mail or parcel, and includes different categories such as documents, merchandise, parts, etc.

[1189] "Item size" refers to the physical size or dimensions of a mail or parcel.

[1190] "Form of delivery" refers to the shape and type of packaging of mail or parcels.

[1191] "Candidate destinations" are delivery destinations suggested for unknown mail or parcels based on past data and current information.

[1192] An "image processing device" is a device that captures images using a camera or sensor, analyzes the images, and extracts the necessary information.

[1193] A "machine learning model" is an algorithm that learns patterns based on past data and makes predictions and classifications for new data.

[1194] "Real-time" refers to a state in which the time from data acquisition to processing and output of results is extremely short, with results being obtained almost instantly.

[1195] As an embodiment of the present invention, a system for streamlining the sorting of mail and parcels with unknown destinations at a logistics center will be described. In this system, workers use smart glasses to scan deliveries and suggest optimal drop-off destinations in real time.

[1196] Hardware and Software Used

[1197] Hardware:

[1198] Smart glasses (e.g. Google Glass)

[1199] Camera (built into smart glasses)

[1200] software:

[1201] OpenCV (image processing library)

[1202] scikit-learn (machine learning library)

[1203] SQLite (database)

[1204] System Configuration

[1205] 1. Image processing device:

[1206] The smart glasses use a built-in camera to scan the package, and the images captured by the camera are processed using OpenCV to identify the sender and the type and size of the package.

[1207] 2. Database:

[1208] SQLite is used to store detailed information about past mailing destinations and deliveries, which makes it possible to suggest mailing destination candidates based on past data.

[1209] 3. Machine learning models:

[1210] Using scikit-learn, we train a KNN (k-Nearest Neighbors) model based on historical data, which then suggests optimal delivery locations in real time based on the sender and the type, size, and format of the parcel.

[1211] Processing flow

[1212] 1. Scan:

[1213] A worker wears smart glasses and scans the package with a camera. The image captured by the camera is processed using OpenCV to identify the sender and the type and size of the package.

[1214] 2. Database Matching:

[1215] The recognized information is checked against a SQLite database to obtain details about past mailing destinations and deliveries.

[1216] 3. Suggestions for possible recipients:

[1217] Using a machine learning model (KNN), the system will suggest the best possible recipient in real time based on the recognized information and past data, and this information will be displayed on the smart glasses' display.

[1218] Examples

[1219] For example, if the sender is "Company A," the delivery item is "Parts Type B" and the size is "Medium," the system will suggest that the item be sent to "Department C" based on past data. This suggestion will be displayed in real time on the smart glasses' display.

[1220] Examples of prompt statements

[1221] "Please suggest the best delivery location for the item whose sender is Company A, the item is a type B part, and the size is medium."

[1222] In this way, the sorting of mail and parcels with unknown destinations can be made more efficient, the risk of misdelivery can be reduced, and the accuracy of the sorting process can be improved.

[1223] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1224] Step 1:

[1225] A user wears smart glasses and scans a package with the camera. The input is an image of the package captured by the camera, and the output is image data. The camera of the smart glasses captures the image of the package and performs image processing using OpenCV. Specifically, the sender, type of package, and size are extracted from the image.

[1226] Step 2:

[1227] The server uses an image processing device to recognize the sender and the type and size of the delivery from the scanned image. The input is the image data acquired in step 1, and the output is the recognized sender, type and size of the delivery. OpenCV is used to analyze the image and extract text information and dimensions.

[1228] Step 3:

[1229] The server uses the information it has learned to match it with a SQLite database. The input is the sender, type of delivery, and size of the delivery, and the output is details of past delivery destinations and deliveries. A database query is performed to obtain details of past delivery destinations and deliveries.

[1230] Step 4:

[1231] The server uses a machine learning model (KNN) to suggest optimal delivery destination candidates in real time. The input is the sender, the type of delivery, the size, and past delivery destination information, and the output is the optimal delivery destination candidate. The KNN model is run using scikit-learn to predict the optimal delivery destination.

[1232] Step 5:

[1233] The server displays the suggested drop-off candidates on the display of the smart glasses. The input is the optimal drop-off candidate, and the output is the information to be displayed on the display of the smart glasses. The drop-off candidates are displayed on the display of the smart glasses in real time and notified to the user.

[1234] In this way, the sorting of mail and parcels with unknown destinations can be made more efficient, the risk of misdelivery can be reduced, and the accuracy of the sorting process can be improved.

[1235] Example 2

[1236] Next, a second embodiment of the second form will be described. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal".

[1237] Sorting mail and parcels with unknown addresses often depends on the work experience and sorting lists of the person in charge, which can be inefficient. In addition, there is a risk that the mail will be delivered to the wrong address because the details of past mailing destinations and deliveries cannot be properly utilized. Furthermore, there is an issue that it is difficult to suggest possible mailing destinations based on the sender and the type, size, and format of the delivery.

[1238] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1239] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for generating possible destinations using a generative AI model. This makes it possible to efficiently sort mail and parcels with unknown addresses and prevent delivery to the wrong destination.

[1240] "Sender" refers to the person or organization sending the mail or package.

[1241] "Delivery" refers to goods or documents sent by mail or courier.

[1242] "Type" indicates the classification of the delivery item, and refers to categories such as documents, parts, and goods.

[1243] "Size" refers to the size of the delivery item, such as small, medium, large, etc.

[1244] "Form" refers to the shape of the delivery item and the state of packaging, such as a box, envelope, or pallet.

[1245] "Potential drop-off locations" refer to locations or departments where a delivery may be dropped off.

[1246] "Previous delivery destination" refers to a location or department where a delivery has previously been delivered.

[1247] "Detailed information" refers to specific data and historical information about the delivery.

[1248] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to analyze data and generate appropriate recipient candidates.

[1249] "Database" refers to a system for storing and managing past mailing history and detailed delivery information.

[1250] "User" refers to the person or organization that operates the system to input data and review / decide on suggested recipients.

[1251] The present invention is a system for streamlining the sorting of mail and parcels with unknown addresses. This system is equipped with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. A specific embodiment of this system will be described below.

[1252] Hardware and Software Used

[1253] Hardware: Servers, terminals (PCs, tablets, etc.)

[1254] Software: Generative AI models (e.g. GPT-4), database management systems (e.g. MySQL), postal and parcel management systems

[1255] System program processing

[1256] 1. Data Entry

[1257] The user uses the terminal to enter data such as sender information and the type, size and format of the delivery item.

[1258] Example: A user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal and clicks the send button.

[1259] 2. Database Reference

[1260] The server uses a database management system (e.g. MySQL) to look up details of past mailings and deliveries.

[1261] Example: The server executes an SQL query and sends the following command to the database: "SELECT FROM mailing history WHERE sender = 'Company A' AND delivery = 'Parts of type B'".

[1262] 3. Generating recipient candidates

[1263] The server uses a generative AI model (e.g. GPT-4) to generate potential recipients based on the input data and past data.

[1264] Example: The server sends a prompt to the generative AI model: "If the sender is Company A and the delivery is a type B part, please suggest an appropriate delivery destination based on past data." The generative AI model responds: "Department C is appropriate."

[1265] 4. Suggestions for possible recipients

[1266] The server transmits the generated candidate mailbox destinations to the terminal and suggests them to the user.

[1267] Example: The server sends a message to the terminal saying, "We suggest posting to Department C." The content of the suggestion is displayed on the terminal screen.

[1268] 5. User confirmation and final decision

[1269] The user checks the proposed destination candidates and decides on the final destination.

[1270] Example: The user checks the proposal displayed on the terminal screen and clicks the "Approve posting to Department C" button. The user's selection is sent to the server, and the final posting destination is confirmed.

[1271] Examples of prompt statements

[1272] If the sender is "Company A", the delivery item is "Type B Parts" and the size is "Medium", please suggest an appropriate drop-off location based on past data.

[1273] In this way, users can efficiently sort mail and parcels that are not addressed to their destination. By utilizing generative AI models and a database management system, the system can accurately suggest possible recipients based on past data.

[1274] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1275] Step 1:

[1276] Data Entry

[1277] The user uses the terminal to enter data such as sender information and the type, size and format of the delivery item.

[1278] Input: sender information (e.g. Company A), type of delivery (e.g. Type B parts), size (e.g. medium size)

[1279] Specific operation: The user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal and clicks the send button.

[1280] Output: The entered data is sent to the server.

[1281] Step 2:

[1282] Database Reference

[1283] The server uses a database management system (e.g. MySQL) to look up details of past mailings and deliveries.

[1284] Input: Data entered by the user (sender information, type of delivery, size)

[1285] Specific operation: The server executes an SQL query and sends the command "SELECT FROM mailing history WHERE sender = 'Company A' AND delivery = 'Parts of type B'" to the database.

[1286] Output: Detailed information about past mailing destinations and deliveries is obtained.

[1287] Step 3:

[1288] Generating recipient candidates

[1289] The server uses a generative AI model (e.g. GPT-4) to generate potential recipients based on the input data and past data.

[1290] Input: Data entered by the user, details of past mailings and deliveries retrieved from a database

[1291] Specific operation: The server sends a prompt to the generative AI model: "If the sender is Company A and the delivery is a type B part, please suggest an appropriate delivery destination based on past data." The generative AI model responds: "Department C is appropriate."

[1292] Output: A list of possible recipients (e.g. Department C) is generated.

[1293] Step 4:

[1294] Suggestions for possible recipients

[1295] The server transmits the generated candidate mailbox destinations to the terminal and suggests them to the user.

[1296] Input: Generated recipient candidates (e.g. Department C)

[1297] Specific operation: The server sends a message to the terminal saying, "We suggest posting to Department C." The content of the suggestion is displayed on the terminal screen.

[1298] Output: The suggestions are displayed to the user.

[1299] Step 5:

[1300] User confirmation and final decision

[1301] The user checks the proposed destination candidates and decides on the final destination.

[1302] Input: Suggested recipient (e.g. Department C)

[1303] Specific operation: The user checks the proposal displayed on the terminal screen and clicks the "Approve submission to Department C" button. The user's selection is sent to the server.

[1304] Output: The final posting destination is confirmed.

[1305] In this way, users can efficiently sort mail and parcels that are not addressed to their destination. By utilizing generative AI models and a database management system, the system can accurately suggest possible recipients based on past data.

[1306] (Application example 2)

[1307] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server", and the headset type terminal 314 will be referred to as a "terminal".

[1308] The conventional method of sorting mail and parcels with unknown addresses requires the worker to check the work experience and sorting list each time, which is inefficient and prone to human error. In addition, there is no system that uses past data to suggest the optimal destination, so there is a need to improve the accuracy and efficiency of the sorting process.

[1309] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1310] In this invention, the server includes a means for suggesting possible post destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible post destinations based on past post destinations and detailed information on the delivery, a means installed in the smartphone for suggesting the optimal post destination based on past data, and a means for saving the history of the suggested post destinations and improving the accuracy of the suggestions from the next time onwards. This makes it possible to significantly improve the efficiency and accuracy of the sorting work for mail and parcels with unknown addresses.

[1311] "Unknown address mail or parcel" refers to mail or parcel that cannot be delivered to a specific address through normal delivery procedures because the address specified by the sender is unclear.

[1312] "Sorting work" refers to the task of classifying mail and parcels based on their destination and content, and then distributing them to the appropriate destination.

[1313] "Sender" refers to the individual or organization sending the mail or parcel.

[1314] "Type of delivery" refers to the category of the contents of a mail or parcel, such as documents, merchandise, parts, etc.

[1315] "Size" refers to the physical size of a mail item or parcel, and is classified as small, medium, large, etc.

[1316] "Format" refers to the shape and type of packaging of a mail or parcel, such as an envelope, box, or pallet.

[1317] "Potential drop-off locations" refer to locations or departments to which undeliverable mail or parcels may be delivered.

[1318] "Previous mailing destination" refers to a location or department to which undeliverable mail or parcels have previously been delivered.

[1319] "Detailed information" refers to specific data and history about mail or parcels.

[1320] A "smartphone" is a type of mobile phone and a multi-function device that can connect to the Internet and install applications.

[1321] "Saving history" means recording information about past mailing destinations and deliveries in a database so that it can be referenced later.

[1322] "Suggestion accuracy" refers to the system's ability to accurately suggest the optimal drop-off destination.

[1323] As an embodiment of the present invention, a system for streamlining the sorting of mail and parcels with unknown destinations is constructed. A specific embodiment of the system is described below.

[1324] System Configuration

[1325] Hardware:

[1326] Smartphone (iOS or Android)

[1327] software:

[1328] Python

[1329] Pandas (Data Processing)

[1330] Scikit-learn (machine learning model)

[1331] Program processing explanation

[1332] Data Loading:

[1333] The server reads past delivery data from a CSV file, which gives information about the sender, type of delivery, size, format, and past delivery destination.

[1334] Split features and labels:

[1335] Based on the data it reads, the server classifies the sender, type of delivery, size, and format as features, and the delivery destination as a label.

[1336] Model training:

[1337] The server trains a machine learning model using K-nearest neighbors (KNN) to improve its ability to suggest optimal drop destinations based on past data.

[1338] Suggested destinations:

[1339] When a user uses their smartphone to input information about a new delivery (sender, type of delivery, size, format), the server uses the trained model to suggest the optimal delivery location.

[1340] Save history:

[1341] The server stores the history of suggested destinations in a database, which improves the accuracy of suggestions from the next time onwards.

[1342] Examples

[1343] For example, if the sender is "Company A," the delivery item is a "Type B part," the size is "medium," and the format is "box," it will suggest posting it to "Department C" based on past data.

[1344] Examples of prompt statements

[1345] Please use the following information to suggest the best place to send your message.

[1346] From: Company A

[1347] Delivery type: B-type parts

[1348] Size: Medium

[1349] Format: Box

[1350] In this way, a system can be realized that improves the efficiency and accuracy of sorting mail and parcels with unknown destinations.

[1351] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1352] Step 1:

[1353] Data Import

[1354] The server reads past delivery data from a CSV file. The path of the CSV file is specified as input. The server reads the CSV file in data frame format using the Pandas library and obtains information on the sender, type of delivery, size, format, and past recipients. A data frame is generated as output.

[1355] Step 2:

[1356] Splitting features and labels

[1357] Based on the data read, the server separates the sender, type, size, and format of the delivery as features, and the recipient as a label. The data frame generated in step 1 is used as input. The server splits the features and labels into separate data frames, in preparation for training the machine learning model. As output, a feature data frame and a label data frame are generated.

[1358] Step 3:

[1359] Model Training

[1360] The server trains a machine learning model using K-nearest neighbors (KNN). As input, the feature and label dataframes generated in step 2 are used. The server trains the KNN model using the Scikit-learn library to improve its ability to suggest optimal drop destinations based on past data. As output, a trained KNN model is generated.

[1361] Step 4:

[1362] Suggested destination

[1363] A user inputs new parcel information (sender, parcel type, size, format) using a smartphone. The parcel information entered by the user is used as input. The server uses the trained KNN model to suggest the optimal delivery destination based on the input information. As output, the suggested delivery destination is displayed to the user.

[1364] Step 5:

[1365] History storage

[1366] The server saves the history of suggested destinations in a database. As input, the proposed destinations in step 4 and the delivery information entered by the user are used. The server records this information in the database to improve the accuracy of future suggestions. As output, the history data is saved in the database.

[1367] In this way, a system can be realized that improves the efficiency and accuracy of sorting mail and parcels with unknown destinations.

[1368] In addition, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1369] "Example 1"

[1370] As an embodiment of the present invention, a system incorporating an emotion engine is provided. The system recognizes the user's emotion and adjusts the candidate destinations based on the emotion. Specifically, if the system recognizes that the user is stressed, the system reduces the number of candidate destinations and simplifies the options. Also, if the system recognizes that the user is relaxed, the system provides more candidate destinations, giving the user more freedom of choice. Furthermore, the emotion engine also adjusts the way in which the candidate destinations are presented based on the user's emotion. For example, if the system recognizes that the user is anxious, the system does not display a list of candidate destinations, but directly presents the most likely destination. This allows the user to make a decision quickly.

[1371] "Example 2"

[1372] As an embodiment of the present invention, a system incorporating an emotion engine is provided. The system recognizes the user's emotion and adjusts the candidate destinations based on the emotion. Specifically, if the system recognizes that the user is stressed, the system reduces the number of candidate destinations and simplifies the options. Also, if the system recognizes that the user is relaxed, the system provides more candidate destinations, giving the user more freedom of choice. Furthermore, the emotion engine also adjusts the way in which the candidate destinations are presented based on the user's emotion. For example, if the system recognizes that the user is anxious, the system does not display a list of candidate destinations, but directly presents the most likely destination. This allows the user to make a decision quickly.

[1373] The process flow of each embodiment will be described below.

[1374] "Example 1"

[1375] Step 1: The system recognizes the user's emotions using an emotion engine, which infers emotions from the user's facial expressions, tone of voice, choice of words, etc.

[1376] Step 2: After the emotion engine recognizes the user's emotion, the system will adjust the suggested destinations based on the emotion. For example, if the system recognizes that the user is stressed, it will reduce the number of suggested destinations and simplify the options.

[1377] Step 3: On the other hand, if the system detects that the user is relaxed, it will provide more possible destinations, giving the user more freedom of choice.

[1378] Step 4: Furthermore, the emotion engine will also adjust the way in which the system presents the possible destinations based on the user's emotions. For example, if the system detects that the user is in a hurry, it will directly present the most likely destination instead of displaying a list of possible destinations, allowing the user to make a decision quickly.

[1379] Example 1

[1380] Next, a description will be given of Example 1 of the embodiment 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal".

[1381] Conventional methods for sorting mail and parcels with unknown addresses are inefficient because they require the person in charge to check the work experience and sorting list each time. In addition, there is a problem that appropriate support cannot be provided to users who feel stressed or impatient because the system cannot respond flexibly to the user's emotional state. This has led to a demand for more efficient sorting work and improved user satisfaction.

[1382] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1383] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on the emotions. This makes it possible to streamline the sorting of mail and parcels with unknown addresses and to respond flexibly to the user's emotional state.

[1384] "Sender" refers to the individual or organization sending the mail or parcel.

[1385] "Delivery" refers to goods or documents sent by mail or courier.

[1386] "Type" is an attribute that indicates the classification of the delivery, and includes, for example, documents, parts, and merchandise.

[1387] "Size" is an attribute that indicates the physical size of a delivery item, and includes, for example, length, width, height, weight, and the like.

[1388] "Form" is an attribute that indicates the shape of the delivery item or the state of packaging, and includes, for example, a box, an envelope, a pallet, and the like.

[1389] "Potential drop-off locations" refers to locations or departments where deliveries may be dropped off.

[1390] "Previous mailbox locations" refers to records of locations or departments where deliveries have been previously mailed.

[1391] "Detailed information" refers to specific data and history about the delivery.

[1392] "Emotion recognition" refers to identifying a user's emotional state by analyzing their facial expressions and tone of voice.

[1393] "Adjust" refers to changing the system's behavior or suggestions to suit specific conditions or circumstances.

[1394] The present invention is a system for streamlining the sorting of mail and parcels with unknown destinations. This system includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on those emotions.

[1395] Hardware and Software Used

[1396] Hardware

[1397] Server: Analyzes and stores data.

[1398] Terminal: Provides a user interface and is equipped with a camera and microphone for recognizing the user's emotions.

[1399] Scanner: Scan postal and courier labels.

[1400] software

[1401] Google Cloud Vision API: Converts paper data into digital data using OCR (optical character recognition) technology.

[1402] MySQL: A database management system that stores analysis results and past submission data.

[1403] Microsoft Azure Emotion API: Recognizes emotions by analyzing the user's facial expressions and tone of voice.

[1404] System Operation

[1405] Data Entry and Analysis

[1406] The server receives data such as sender information for mail and parcels, and the type, size, and format of the delivery. Specifically, it scans the mail and parcel labels and converts them from paper to digital data using Google Cloud Vision API. The server analyzes the received data and identifies the sender and the type, size, and format of the delivery. The analysis results are stored in a MySQL database.

[1407] Suggestions for possible recipients

[1408] The server searches a database of past delivery destinations based on the analysis results and suggests the most appropriate delivery destination. For example, if the sender is "Company A" and the delivery item is "Type B parts," it suggests delivery to "Department C" based on past data. Also, if the size of the delivery item falls within a certain range, it suggests delivery to a specific location.

[1409] Emotion engine adjustment

[1410] The device uses a camera and microphone to analyze the user's facial expressions and tone of voice to recognize the user's emotions. It uses Microsoft Azure's Emotion API. If the server recognizes that the user is stressed, it reduces the number of possible delivery candidates and simplifies the options. Conversely, if the server recognizes that the user is relaxed, it provides more possible delivery candidates.

[1411] Suggestion of possible recipients

[1412] The server sends the adjusted drop-off candidates to the terminal and presents them to the user. If the server detects that the user is in a hurry, it will directly present the most likely drop-off destination, allowing the user to make a decision quickly.

[1413] Examples of concrete examples and prompts

[1414] Examples

[1415] When a user receives a "Part B" from "Company A", the system will suggest dropping it off at "Division C" based on past data. If the system detects that the user is stressed, it will only suggest dropping it off at "Division C".

[1416] Examples of prompt statements

[1417] "We have received part B from company A. Please suggest the best delivery destination based on past data. Also, please adjust the suggestions based on user sentiment."

[1418] The above is an embodiment of the present invention.

[1419] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1420] Step 1:

[1421] Data Entry

[1422] The server receives data such as sender information for mail or parcels, and the type, size, and format of the delivery. As input, it scans the label of the mail or parcel to obtain image data. In concrete terms, the server obtains image data from the scanner and calls the Google Cloud Vision API to extract text information from the image data. As output, it obtains the extracted text information.

[1423] Step 2:

[1424] 4. Data Analysis

[1425] The server analyzes the text information obtained in step 1 to identify the sender, type of delivery, size, and format. As input, it receives the extracted text information. In concrete terms, the server analyzes the text information to identify the sender, type of delivery, size, and format. As output, it obtains the analysis results, which are stored in a MySQL database.

[1426] Step 3:

[1427] Suggestions for possible recipients

[1428] The server searches the database of past destinations based on the analysis results obtained in step 2, and proposes the most suitable destination candidates. It receives the analysis results as input. In concrete terms, the server queries the MySQL database to retrieve the past destination data. As output, it obtains a list of optimal destination candidates.

[1429] Step 4:

[1430] Emotion engine adjustment

[1431] To recognize the user's emotions, the device uses a camera and microphone to analyze the user's facial expressions and tone of voice. It receives the user's facial expression and voice data as input. Specifically, the device calls Microsoft Azure's Emotion API to analyze the user's emotions. The output is the user's emotional state. The server adjusts the number of possible recipients based on the user's emotional state.

[1432] Step 5:

[1433] Suggestion of possible recipients

[1434] The server sends the adjusted list of destination candidates in step 4 to the terminal and presents them to the user. As input, it receives the adjusted list of destination candidates. In concrete terms, the server sends the adjusted destination candidates to the terminal, and the terminal displays the destination candidates on the display. As output, the destination candidates presented to the user are obtained.

[1435] (Application example 1)

[1436] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server", and the headset type terminal 314 will be referred to as a "terminal".

[1437] The conventional method for sorting mail and parcels with unknown addresses requires the person in charge to check the work experience and sorting list each time, which makes it inefficient and prone to errors. In addition, there are many situations that require flexible responses based on the user's emotions, and there was no system that could handle this.

[1438] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1439] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the number of possible destinations and the presentation method based on the emotions. This makes it possible to streamline the work of sorting mail and parcels with unknown addresses and to respond flexibly to the user's emotions.

[1440] "Sender" refers to the person or organization sending the mail or package.

[1441] "Delivery" refers to goods or documents sent by mail or courier.

[1442] "Type" is an attribute that indicates the classification of the delivery, such as documents, parts, or merchandise.

[1443] "Size" is an attribute that indicates the size or dimensions of the delivery item.

[1444] "Form" is an attribute that indicates the shape of the delivery and the state of packaging.

[1445] "Possible drop-off locations" refers to possible locations where deliveries may be dropped off or potential recipients.

[1446] "Previous delivery destination" refers to a location or recipient where a delivery has previously been delivered.

[1447] "Detailed information" refers to specific data and historical information about the delivery.

[1448] "Emotion recognition" refers to detecting and understanding a user's emotional state.

[1449] "Emotion-based adjustment" refers to changing the system's behavior or presentation in response to recognized emotions.

[1450] A system for implementing the present invention includes a server, a smartphone, an emotion recognition engine, and a machine learning model. Specific embodiments are described below.

[1451] System Configuration

[1452] Hardware

[1453] Server: Stores and processes data.

[1454] Smartphones: Used as a user interface and receive user input.

[1455] Emotion recognition engine: Recognize user emotions in real time.

[1456] Machine learning model: A model for suggesting possible recipients (e.g., KNN model).

[1457] software

[1458] Python: Used to implement the program.

[1459] scikit-learn: Used to train and predict machine learning models.

[1460] emotion_recognition: A library of emotion recognition engines.

[1461] Data processing and calculation

[1462] server

[1463] The server stores past delivery destination data and delivery details, retrieves them when needed, and also trains machine learning models to suggest delivery destinations based on user input.

[1464] Smartphone

[1465] The smartphone receives input from the user (sender, type, size, and format of the delivery item) and sends it to the server. It also recognizes the user's emotions in real time through an emotion recognition engine and sends that information to the server.

[1466] Emotion Recognition Engine

[1467] The emotion recognition engine recognizes the user's emotions from their facial expressions and voice, and returns the results to the smartphone, which then adjusts the number of possible recipients and the way they are presented according to the user's emotional state.

[1468] Machine learning models

[1469] A machine learning model (e.g., KNN model) is trained based on past data and suggests optimal delivery candidates based on user input data. It receives information from an emotion recognition engine and adjusts the suggestions according to the user's emotions.

[1470] Examples

[1471] For example, if the sender is "Company A," the delivery is "Type B parts," and the size is "medium," the system will suggest dropping it off at "Department C" based on past data. If the user is feeling stressed, the system will narrow down the suggestions to one, and if the user is relaxed, it will display up to five suggestions.

[1472] Examples of prompt statements

[1473] Suggest possible delivery destinations for a "medium" size package with a "Type B part" from "Company A". If the user is stressed, narrow down the suggestions to one, and if the user is relaxed, display up to five suggestions.

[1474] In this way, a system is provided that can streamline the sorting of mail and parcels with unknown destinations and can flexibly respond to the user's emotions.

[1475] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1476] Step 1:

[1477] The user uses a smartphone to input the sender, type of delivery, size, and format. The input data is sent from the smartphone to the server. The input data includes the sender name, type of delivery, size, and format.

[1478] Step 2:

[1479] The smartphone uses an emotion recognition engine to recognize the user's emotions in real time. The emotion recognition engine analyzes the user's facial expressions and voice data to detect their emotional state (e.g., stress, relaxation). The detected emotion data is sent from the smartphone to a server.

[1480] Step 3:

[1481] The server retrieves relevant data from a database of past mailing destinations based on the received input data (sender, type of delivery, size, format) and emotion data. The server compares the past mailing destination data with the detailed delivery information to generate optimal mailing destination candidates.

[1482] Step 4:

[1483] The server uses a machine learning model (e.g., KNN model) to suggest possible destinations based on the input data. The machine learning model is trained based on past data and analyzes the input data to predict the best possible destinations. The predicted destination candidates are sent from the server to the smartphone.

[1484] Step 5:

[1485] The server adjusts the number and presentation of possible recipients based on the emotion data. For example, if the user is stressed, the server reduces the number of possible recipients and presents only one most likely recipient. If the user is relaxed, the server presents multiple possible recipients (e.g., five).

[1486] Step 6:

[1487] The smartphone displays the candidate destinations received from the server to the user. The user selects the most suitable destination from the presented candidates. The selected destination is sent from the smartphone to the server.

[1488] Step 7:

[1489] The server records the destination selected by the user and updates the database to improve the accuracy of the suggestions for the next time, allowing the system to continuously learn and improve the accuracy of the suggestions.

[1490] In this way, a system is provided that can streamline the sorting of mail and parcels with unknown destinations and can flexibly respond to the user's emotions.

[1491] Example 2

[1492] Next, a second embodiment of the second form will be described. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal".

[1493] Conventional methods for sorting mail and parcels with unknown addresses rely on the work experience and sorting lists of the person in charge, which makes them inefficient and prone to errors. In addition, there is also the problem that the system is unable to respond flexibly to the user's emotional state, making it difficult to provide appropriate support to users who feel stressed or impatient. Furthermore, there are insufficient suggestions for delivery destinations based on the sender and the type, size, and format of the delivery, making it difficult to quickly find the optimal delivery destination.

[1494] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1495] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on the emotions. This makes it possible to streamline the sorting of mail and parcels with unknown destinations and reduce errors. In addition, since the server can flexibly respond to the user's emotional state, it reduces the user's stress and impatience and allows the user to quickly find the optimal destination.

[1496] "Sender" refers to the person or organization sending the mail or package.

[1497] "Delivery" refers to goods or documents sent by mail or courier.

[1498] "Type" refers to the classification or category of the delivery, such as documents, parts, merchandise, etc.

[1499] "Size" refers to the size or dimensions of the delivery item, and includes, for example, small, medium, large, etc.

[1500] "Format" refers to the shape or packaging of the delivery, such as a box, envelope, or pallet.

[1501] "Potential delivery destinations" refers to potential locations or departments where deliveries may be dropped off.

[1502] "Previous delivery destination" refers to a location or department where a delivery has previously been delivered.

[1503] "Detailed information" refers to specific data and historical information about the delivery.

[1504] "Emotion recognition engine" refers to software or hardware for analyzing and recognizing a user's emotional state.

[1505] A "generative AI model" refers to a model that uses artificial intelligence to analyze data and generate optimal results.

[1506] A "prompt" refers to an instruction or question that is input into a generative AI model.

[1507] The present invention is a system for streamlining the sorting of mail and parcels with unknown addresses, and includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery item, a means for suggesting possible destinations based on past delivery destinations and detailed information about the delivery item, and a means for recognizing the user's emotions and adjusting the possible destinations based on those emotions.

[1508] Hardware and Software Used

[1509] Hardware: Servers, user devices (PCs, tablets, smartphones)

[1510] Software: Generative AI models, emotion recognition engine, database management system (DBMS)

[1511] Specific operation of the system

[1512] 1. Data Entry:

[1513] The user uses the terminal to enter information such as the sender and the type, size, and format of the delivery.

[1514] Specific operation: The user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal.

[1515] 2. Data Analysis:

[1516] The server receives the entered data and compares it with its historical database.

[1517] Specific operation: The server uses a database management system (DBMS) to search for data that "Parts of type B from Company A were previously mailed to Department C."

[1518] 3. Suggestions for possible recipients:

[1519] The server uses a generative AI model to suggest optimal delivery candidates.

[1520] Specific operation: The server provides input data as prompts to the generative AI model to generate optimal delivery candidates.

[1521] Example: The server inputs the prompt "Sender: Company A, Delivery: Type B part, Size: Medium" into the generative AI model and outputs "Department C" as a possible recipient.

[1522] 4. Emotion recognition:

[1523] The terminal uses an emotion recognition engine to analyze the user's emotions in real time while the user is operating the terminal.

[1524] Specific behavior: Uses the device's camera and microphone to analyze the user's facial expressions and tone of voice to determine their emotional state.

[1525] Example: The device captures the user's face with a camera and an emotion recognition engine detects "stress."

[1526] 5. Adjustment of possible recipients:

[1527] The server adjusts the number of possible recipients and the presentation method based on the user's feelings.

[1528] What it does: The server receives data from the emotion recognition engine and refines the list of potential recipients.

[1529] Example: The server receives information that "the user is feeling stressed" and narrows down the possible recipients to only "Department C".

[1530] Examples

[1531] Example of input prompt:

[1532] "If the sender is Company A and the delivery item is Type B parts, please suggest the best delivery location based on past data."

[1533] "If users are feeling stressed, present them with fewer possible recipients."

[1534] In this way, the system suggests optimal candidates for recipients based on the user's input data and emotional state.

[1535] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1536] Step 1:

[1537] Data Entry

[1538] The user uses the terminal to enter information such as the sender and the type, size, and format of the delivery.

[1539] Input: sender, delivery type, size, format

[1540] Specific operation: The user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal.

[1541] Output: The entered data is sent to the server.

[1542] Step 2:

[1543] 4. Data Analysis

[1544] The server receives the entered data and compares it with its historical database.

[1545] Input: Data sent by the user, such as sender, type of delivery, size, and format

[1546] Specific operation: The server uses a database management system (DBMS) to search for data that "Parts of type B from Company A were previously mailed to Department C."

[1547] Output: The matching result is the past mailing destination data.

[1548] Step 3:

[1549] Suggestions for possible recipients

[1550] The server uses a generative AI model to suggest optimal delivery candidates.

[1551] Input: Past mailing destination data obtained as a result of matching, and data input by users

[1552] Specific operation: The server provides input data as prompts to the generative AI model to generate optimal delivery candidates.

[1553] Output: A list of potential recipients from the generative AI model.

[1554] Step 4:

[1555] emotion recognition

[1556] The terminal uses an emotion recognition engine to analyze the user's emotions in real time while the user is operating the terminal.

[1557] Input: User facial expressions and tone of voice

[1558] Specific behavior: Uses the device's camera and microphone to analyze the user's facial expressions and tone of voice to determine their emotional state.

[1559] Output: Emotion state data from the emotion recognition engine.

[1560] Step 5:

[1561] Adjustment of possible recipients

[1562] The server adjusts the number of possible recipients and the presentation method based on the user's feelings.

[1563] Input: Emotion state data from an emotion recognition engine, a list of recipient candidates from a generative AI model

[1564] What it does: The server receives data from the emotion recognition engine and refines the list of potential recipients.

[1565] Output: The refined list of possible recipients is presented to the user.

[1566] In this way, the system suggests optimal candidates for recipients based on the user's input data and emotional state.

[1567] (Application example 2)

[1568] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server", and the headset type terminal 314 will be referred to as a "terminal".

[1569] Sorting mail and parcels with unknown addresses is difficult because it is difficult to find the appropriate place to drop them off based on the sender and the type, size, and format of the parcel, and because it requires the staff member to check the work experience and sorting list each time.Furthermore, flexible responses according to the user's emotional state are required, but conventional systems were unable to achieve this.

[1570] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1571] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information of the delivery, a means for recognizing the user's emotions and adjusting the possible destinations based on the emotions, a means for scanning the delivery with a camera in the smart glasses and recognizing the sender and the type, size, and format of the delivery, and a means for presenting the optimal possible destinations based on the recognized information, past data, and the user's emotions. This makes it possible to efficiently and flexibly sort mail and parcels with unknown addresses.

[1572] A "sender" is a person or organization that sends a letter or package.

[1573] A "delivery" is an item sent by mail or courier.

[1574] "Type" is an attribute that indicates the classification of the delivery.

[1575] "Size" is an attribute that indicates the size of the delivery item.

[1576] "Form" is an attribute that indicates the shape of the delivery item and the type of packaging.

[1577] "Potential delivery destinations" are potential locations or departments where deliveries may be delivered.

[1578] A "previous delivery destination" is a location or department where a delivery has been previously delivered.

[1579] "Detailed information" refers to specific data and history information regarding the delivery.

[1580] "User emotion" refers to the psychological state of a person using a system.

[1581] "Smart glasses" are glasses-type devices that have built-in cameras and displays and have the ability to display information.

[1582] A "camera" is a device that captures images or videos.

[1583] "Recognition" refers to the system analyzing input information and determining specific attributes or conditions.

[1584] "Data" is a collection of information that is processed by a system.

[1585] "Presentation" means that the system displays information to the user.

[1586] A system for implementing the present invention includes smart glasses, a camera, an emotion recognition engine, a database, and a display for presenting possible recipients. The specific configuration and operation of the system will be described below.

[1587] System Configuration

[1588] 1. Smart glasses: The smart glasses are equipped with a camera and a display and are worn by workers. The camera scans the package and the display shows possible recipients.

[1589] 2. Camera: The camera built into the smart glasses takes pictures of the package and provides data to recognize the sender, package type, size and format.

[1590] 3. Emotion Recognition Engine: The emotion recognition engine scans the user's face and recognizes their emotions, allowing it to tailor the recipient suggestions to suit the user's psychological state.

[1591] 4. Database: The database stores detailed information on past mailing destinations and deliveries, and uses this data to suggest possible mailing destinations.

[1592] 5. A means for presenting possible recipients: A means for presenting optimal recipient candidates based on recognized information, past data, and user sentiment.

[1593] System Operation

[1594] 1. Parcel scanning: A worker wears smart glasses and scans the parcel with a camera, which captures an image of the parcel and identifies the sender, type, size and format of the parcel.

[1595] 2. Database query: Based on the recognized information, detailed information on past mailing destinations and deliveries is obtained from the database.

[1596] 3. Emotion Recognition: The camera in the smart glasses scans the user's face and the emotion recognition engine recognizes the user's emotions.

[1597] 4. Suggestion of possible recipients: Suggest optimal recipients based on recognized information, past data, and user emotions. If the user is stressed, the number of possible recipients will be reduced, and if the user is relaxed, more possible recipients will be provided.

[1598] 5. Presentation of possible drop-off destinations: Potential drop-off destinations are displayed on the smart glasses display, allowing the user to make a quick decision.

[1599] Examples

[1600] When a logistics center worker wears smart glasses and scans an unknown delivery, the system automatically recognizes the sender and the type, size, and format of the delivery, and suggests the best possible delivery destination based on past data and the user's emotions. For example, if the sender is "Company A" and the delivery is "Type B parts," the system will suggest delivery to "Department C" based on past data.

[1601] Examples of prompt statements

[1602] "If the sender is Company A and the delivery is for Type B parts, suggest the best possible delivery destinations based on past data. Also, if the user is stressed, provide fewer delivery destinations, and if the user is relaxed, provide more delivery destinations."

[1603] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1604] Step 1:

[1605] The user puts on the smart glasses and scans the delivery with the camera.

[1606] Input: Image of delivery

[1607] Output: Image data

[1608] Specific operation: The camera in the smart glasses takes a picture of the delivery item and captures the image data.

[1609] Step 2:

[1610] The server analyzes the image data and identifies the sender and the type, size and format of the delivery.

[1611] Input: Image data

[1612] Output: sender, delivery type, size, format information

[1613] What it does: Uses image recognition algorithms to extract sender, package type, size and format from image data.

[1614] Step 3:

[1615] Based on the information recognized by the server, detailed information about past mailing destinations and deliveries is retrieved from the database.

[1616] Input: sender, delivery type, size, and format information

[1617] Output: Details of past mailing destinations and deliveries

[1618] What it does: Executes database queries to retrieve details about past mailings and deliveries.

[1619] Step 4:

[1620] The user scans their face with the smart glasses' camera, and the emotion recognition engine recognizes the user's emotions.

[1621] Input: User's face image

[1622] Output: The user's emotional state.

[1623] What it does: It uses emotion recognition algorithms to analyze the user's emotional state from facial images.

[1624] Step 5:

[1625] The server suggests optimal candidates for delivery destinations based on the information recognized by the server, past data, and user sentiment.

[1626] Input: sender, delivery type, size, format information, past mailing destinations and delivery details, user emotional state

[1627] Output: List of possible recipients

[1628] What it does: It uses a generative AI model to analyze input data and generate optimal delivery candidates.

[1629] Step 6:

[1630] The server displays possible delivery destinations on the smart glasses' display.

[1631] Input: List of possible recipients

[1632] Output: Possible recipients displayed on the smart glasses display

[1633] Specific operation: The list of possible recipients is sent to the smart glasses display and displayed to the user.

[1634] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1635] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1636] Another example of generative AI is Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1637] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1638] [Fourth embodiment]

[1639] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1640] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1641] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a wide area network (WAN) and / or a local area network (LAN).

[1642] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. In addition, the microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1643] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[1644] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[1645] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[1646] The control target 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, legs, etc. The posture and behavior of the robot 414 are controlled by controlling the motors of the arms, hands, legs, etc. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1647] Fig. 8 shows an example of main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1648] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1649] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1650] In the robot 414, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1651] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1652] "Example 1"

[1653] The system of the present invention is for sorting mail and parcels with unknown addresses. Specifically, it is provided with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. For example, if the sender is "Company A" and the delivery is "Parts Type B," it is suggested to post the delivery to "Department C" based on past data. In addition, if the size of the delivery falls within a specific range, it is possible to suggest possible destinations according to various conditions, such as suggesting posting to a specific location.

[1654] "Example 2"

[1655] The system of the present invention is for sorting mail and parcels with unknown addresses. Specifically, it is provided with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. For example, if the sender is "Company A" and the delivery is "Parts Type B," it is suggested to post the delivery to "Department C" based on past data. In addition, if the size of the delivery falls within a specific range, it is possible to suggest possible destinations according to various conditions, such as suggesting posting to a specific location.

[1656] The process flow of each embodiment will be described below.

[1657] "Example 1"

[1658] Step 1: The system receives information about the bounced mail or parcel, including the sender, type, size, and format of the parcel.

[1659] Step 2: Next, the system will suggest possible delivery destinations based on the received information. Specifically, it will suggest possible delivery destinations based on the sender and the type, size and format of the delivery.

[1660] Step 3: The system then suggests possible delivery destinations based on past delivery destinations and detailed information about the delivery item. For example, if the sender is "Company A" and the delivery item is "Parts Type B," the system will suggest delivery to "Department C" based on past data.

[1661] Step 4: Finally, the system presents the proposed destination candidates to the user, who can select the most suitable destination from the presented destination candidates.

[1662] Example 1

[1663] Next, a description will be given of Example 1 of the embodiment 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal".

[1664] Sorting mail and parcels with unknown addresses requires the staff to check their work experience and sorting list each time, which is inefficient and increases the risk of misdelivery. There is also the problem that it is difficult to quickly find the appropriate delivery destination based on the sender and the type, size, and format of the delivery. To solve these issues, a system is needed that utilizes past data to efficiently and accurately suggest delivery destination candidates.

[1665] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1666] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, a means for generating possible destinations using a generative AI model, a means for transmitting data entered by a user to the server, a means for the server to acquire past data from a database, a means for the server to transmit the proposed results to a terminal, and a means for the user to check the proposed results and select an appropriate destination. This makes it possible to efficiently and accurately sort mail and parcels with unknown addresses.

[1667] "Sender" refers to the person or organization sending the mail or package.

[1668] "Delivery" refers to items sent by mail or courier service.

[1669] "Type" is an attribute that indicates the classification of the delivery.

[1670] "Size" is an attribute that indicates the size of the delivery item.

[1671] "Form" is an attribute that indicates the shape of the delivery item and the type of packaging.

[1672] "Possible delivery destinations" refer to locations or departments where deliveries may be dropped off.

[1673] "Previous mailbox location" refers to a location or department where a delivery has been previously mailed.

[1674] "Detailed information" refers to specific data and history about the delivery.

[1675] A "generative AI model" is a model that uses artificial intelligence to analyze data and generate potential recipients.

[1676] A "user" is a person who operates the system to enter postal or courier information.

[1677] A "server" is a computer system that processes and stores data.

[1678] A "terminal" is a computer that includes input and display devices operated by a user.

[1679] A "database" is a system for storing detailed information about past mailing destinations and deliveries.

[1680] "Suggested results" refer to the candidate recipients generated by the generative AI model.

[1681] This invention is a system for efficiently and accurately sorting mail and parcels with unknown destinations. This system is realized using a server, a terminal, a generative AI model, and a database management system (DBMS).

[1682] Hardware and Software Used

[1683] Hardware: Servers, terminals

[1684] Software: Generative AI models, Database Management System (DBMS)

[1685] System Configuration

[1686] 1. User enters postal or courier information:

[1687] The user inputs information about the mail or parcel they received, such as the sender, type, size, and format of the parcel, into the input form on the terminal. For example, the sender is "Company A," the parcel is "Type B parts," the size is "medium," and the format is "box."

[1688] 2. The device sends the input data to the server:

[1689] The terminal sends the data entered by the user to the server, specifically, using an HTTP request to send information such as the sender, type of delivery, size, format, etc.

[1690] 3. The server retrieves historical data from the database:

[1691] The server uses a database management system (DBMS) to retrieve past mailing destinations and delivery details. For example, it executes an SQL query to retrieve past mailing destination data for "Parts of type B" from "Company A."

[1692] 4. The server uses the generative AI model to generate possible recipients:

[1693] The server inputs the acquired past data into the generative AI model to generate candidates for recipients. Specifically, it inputs the following prompt sentence to the generative AI model:

[1694] If the sender is "Company A" and the delivery is "Parts Type B", which department would you suggest to post it to based on past data?

[1695] Based on this prompt, the generative AI model suggests sending the item to "Division C."

[1696] 5. The server sends the proposal results to the device:

[1697] The server sends the candidate mailing destinations generated by the generative AI model to the terminal. Specifically, it uses an HTTP response to send the result proposing mailing to "Department C" to the terminal.

[1698] 6. User reviews the suggestions and selects the appropriate destination:

[1699] The user checks the proposed results displayed on the terminal. For example, if the proposal is to send the item to "Division C," the user can accept the proposal or select another destination. The destination selected by the user is recorded in the system and used as reference data for the future.

[1700] In this way, users can efficiently find the correct drop-off location for undelivered mail or parcels, improving the efficiency of sorting operations and reducing the risk of misdelivery.

[1701] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1702] Step 1:

[1703] A user enters postal or courier information.

[1704] The user inputs information about the mail or parcel they received, such as the sender, type of item, size, and format, into the input form on the terminal. For example, the sender is "Company A," the item is "Type B parts," the size is "medium," and the format is "box." The input data is temporarily stored in the terminal's memory.

[1705] Step 2:

[1706] The terminal transmits the input data to the server.

[1707] The terminal sends the data entered by the user to the server using an HTTP request. Specifically, it sends JSON-formatted data including information such as the sender, type of delivery, size, and format to the server. The server receives and analyzes this data.

[1708] Step 3:

[1709] The server retrieves historical data from the database.

[1710] The server uses a database management system (DBMS) to retrieve past mailing destinations and detailed delivery information. For example, it executes an SQL query to retrieve past mailing destination data for "Part B" from "Company A." The retrieved data is stored in the server's memory.

[1711] Step 4:

[1712] The server uses the generative AI model to generate candidate recipients.

[1713] The server inputs the acquired past data into the generative AI model to generate candidates for recipients. Specifically, it inputs the following prompt sentence to the generative AI model:

[1714] If the sender is "Company A" and the delivery is "Parts Type B", which department would you suggest to post it to based on past data?

[1715] Based on this prompt, the generative AI model suggests sending the mail to "Division C." The generated mail destination candidates are stored in the server's memory.

[1716] Step 5:

[1717] The server transmits the proposal results to the terminal.

[1718] The server sends the candidate mailing destinations generated by the generative AI model to the terminal using an HTTP response. Specifically, it sends JSON-formatted data including the results of suggesting mailing to "Department C" to the terminal. The terminal receives and analyzes this data.

[1719] Step 6:

[1720] The user checks the suggestions and selects the appropriate destination.

[1721] The user checks the proposal results displayed on the terminal. For example, if the proposal is to send the item to "Division C," the user can accept the proposal or select another destination. The destination selected by the user is sent from the terminal to the server and recorded in the system. This will be used as reference data in the future.

[1722] (Application example 1)

[1723] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal".

[1724] The conventional method of sorting mail and parcels with unknown addresses relies on the experience of the person in charge and on sorting lists, which results in low efficiency and a high risk of misdelivery. In addition, there is a problem of reduced accuracy in sorting work because there are insufficient suggestions for the delivery destination based on the sender and the type, size, and format of the delivery.

[1725] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means. In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery item, a means for suggesting possible destinations based on past destinations and detailed information of the delivery item, a means for scanning the delivery item using an image processing device and recognizing the sender and the type and size of the delivery item, and a means for suggesting optimal possible destinations in real time using a machine learning model. This makes it possible to streamline the sorting of mail and parcels with unknown destinations, reduce the risk of misdelivery, and improve the accuracy of the sorting work.

[1726] "Undelivered mail or parcel" refers to mail or parcel whose delivery destination cannot be identified using normal delivery procedures because the recipient's address or name is unclear.

[1727] "Sorting" refers to the task of classifying mail and parcels according to their destination and content, and then distributing them to the appropriate destination.

[1728] A "sender" is an individual or organization that sends mail or parcels.

[1729] "Type of delivery" refers to the classification of the contents of a mail or parcel, and includes different categories such as documents, merchandise, parts, etc.

[1730] "Item size" refers to the physical size or dimensions of a mail or parcel.

[1731] "Form of delivery" refers to the shape and type of packaging of mail or parcels.

[1732] "Candidate destinations" are delivery destinations suggested for unknown mail or parcels based on past data and current information.

[1733] An "image processing device" is a device that captures images using a camera or sensor, analyzes the images, and extracts the necessary information.

[1734] A "machine learning model" is an algorithm that learns patterns based on past data and makes predictions and classifications for new data.

[1735] "Real-time" refers to a state in which the time from data acquisition to processing and output of results is extremely short, with results being obtained almost instantly.

[1736] As an embodiment of the present invention, a system for streamlining the sorting of mail and parcels with unknown destinations at a logistics center will be described. In this system, workers use smart glasses to scan deliveries and suggest optimal drop-off destinations in real time.

[1737] Hardware and Software Used

[1738] Hardware:

[1739] Smart glasses (e.g. Google Glass)

[1740] Camera (built into smart glasses)

[1741] software:

[1742] OpenCV (image processing library)

[1743] scikit-learn (machine learning library)

[1744] SQLite (database)

[1745] System Configuration

[1746] 1. Image processing device:

[1747] The smart glasses use a built-in camera to scan the package, and the images captured by the camera are processed using OpenCV to identify the sender and the type and size of the package.

[1748] 2. Database:

[1749] SQLite is used to store detailed information about past mailing destinations and deliveries, which makes it possible to suggest mailing destination candidates based on past data.

[1750] 3. Machine learning models:

[1751] Using scikit-learn, we train a KNN (k-Nearest Neighbors) model based on historical data, which then suggests optimal delivery locations in real time based on the sender and the type, size, and format of the parcel.

[1752] Processing flow

[1753] 1. Scan:

[1754] A worker wears smart glasses and scans the package with a camera. The image captured by the camera is processed using OpenCV to identify the sender and the type and size of the package.

[1755] 2. Database Matching:

[1756] The recognized information is checked against a SQLite database to obtain details about past mailing destinations and deliveries.

[1757] 3. Suggestions for possible recipients:

[1758] Using a machine learning model (KNN), the system will suggest the best possible recipient in real time based on the recognized information and past data, and this information will be displayed on the smart glasses' display.

[1759] Examples

[1760] For example, if the sender is "Company A," the delivery item is "Parts Type B" and the size is "Medium," the system will suggest that the item be sent to "Department C" based on past data. This suggestion will be displayed in real time on the smart glasses' display.

[1761] Examples of prompt statements

[1762] "Please suggest the best delivery location for the item whose sender is Company A, the item is a type B part, and the size is medium."

[1763] In this way, the sorting of mail and parcels with unknown destinations can be made more efficient, the risk of misdelivery can be reduced, and the accuracy of the sorting process can be improved.

[1764] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1765] Step 1:

[1766] A user wears smart glasses and scans a package with the camera. The input is an image of the package captured by the camera, and the output is image data. The camera of the smart glasses captures the image of the package and performs image processing using OpenCV. Specifically, the sender, type of package, and size are extracted from the image.

[1767] Step 2:

[1768] The server uses an image processing device to recognize the sender and the type and size of the delivery from the scanned image. The input is the image data acquired in step 1, and the output is the recognized sender, type and size of the delivery. OpenCV is used to analyze the image and extract text information and dimensions.

[1769] Step 3:

[1770] The server uses the information it has learned to match it with a SQLite database. The input is the sender, type of delivery, and size of the delivery, and the output is details of past delivery destinations and deliveries. A database query is performed to obtain details of past delivery destinations and deliveries.

[1771] Step 4:

[1772] The server uses a machine learning model (KNN) to suggest optimal delivery destination candidates in real time. The input is the sender, the type of delivery, the size, and past delivery destination information, and the output is the optimal delivery destination candidate. The KNN model is run using scikit-learn to predict the optimal delivery destination.

[1773] Step 5:

[1774] The server displays the suggested drop-off candidates on the display of the smart glasses. The input is the optimal drop-off candidate, and the output is the information to be displayed on the display of the smart glasses. The drop-off candidates are displayed on the display of the smart glasses in real time and notified to the user.

[1775] In this way, the sorting of mail and parcels with unknown destinations can be made more efficient, the risk of misdelivery can be reduced, and the accuracy of the sorting process can be improved.

[1776] Example 2

[1777] Next, a second embodiment of the second embodiment will be described. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal".

[1778] Sorting mail and parcels with unknown addresses often depends on the work experience and sorting lists of the person in charge, which can be inefficient. In addition, there is a risk that the mail will be delivered to the wrong address because the details of past mailing destinations and deliveries cannot be properly utilized. Furthermore, there is an issue that it is difficult to suggest possible mailing destinations based on the sender and the type, size, and format of the delivery.

[1779] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1780] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for generating possible destinations using a generative AI model. This makes it possible to efficiently sort mail and parcels with unknown addresses and prevent delivery to the wrong destination.

[1781] "Sender" refers to the person or organization sending the mail or package.

[1782] "Delivery" refers to goods or documents sent by mail or courier.

[1783] "Type" indicates the classification of the delivery item, and refers to categories such as documents, parts, and goods.

[1784] "Size" refers to the size of the delivery item, such as small, medium, large, etc.

[1785] "Form" refers to the shape of the delivery item and the state of packaging, such as a box, envelope, or pallet.

[1786] "Potential drop-off locations" refer to locations or departments where a delivery may be dropped off.

[1787] "Previous delivery destination" refers to a location or department where a delivery has previously been delivered.

[1788] "Detailed information" refers to specific data and historical information about the delivery.

[1789] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to analyze data and generate appropriate recipient candidates.

[1790] "Database" refers to a system for storing and managing past mailing history and detailed delivery information.

[1791] "User" refers to the person or organization that operates the system to input data and review / decide on suggested recipients.

[1792] The present invention is a system for streamlining the sorting of mail and parcels with unknown addresses. This system is equipped with a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, and a means for suggesting possible destinations based on past destinations and detailed information on the delivery. A specific embodiment of this system will be described below.

[1793] Hardware and Software Used

[1794] Hardware: Servers, terminals (PCs, tablets, etc.)

[1795] Software: Generative AI models (e.g. GPT-4), database management systems (e.g. MySQL), postal and parcel management systems

[1796] System program processing

[1797] 1. Data Entry

[1798] The user uses the terminal to enter data such as sender information and the type, size and format of the delivery item.

[1799] Example: A user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal and clicks the send button.

[1800] 2. Database Reference

[1801] The server uses a database management system (e.g. MySQL) to look up details of past mailings and deliveries.

[1802] Example: The server executes an SQL query and sends the following command to the database: "SELECT FROM mailing history WHERE sender = 'Company A' AND delivery = 'Parts of type B'".

[1803] 3. Generating recipient candidates

[1804] The server uses a generative AI model (e.g. GPT-4) to generate potential recipients based on the input data and past data.

[1805] Example: The server sends a prompt to the generative AI model: "If the sender is Company A and the delivery is a type B part, please suggest an appropriate delivery destination based on past data." The generative AI model responds: "Department C is appropriate."

[1806] 4. Suggestions for possible recipients

[1807] The server transmits the generated candidate mailbox destinations to the terminal and suggests them to the user.

[1808] Example: The server sends a message to the terminal saying, "We suggest posting to Department C." The content of the suggestion is displayed on the terminal screen.

[1809] 5. User confirmation and final decision

[1810] The user checks the proposed destination candidates and decides on the final destination.

[1811] Example: The user checks the proposal displayed on the terminal screen and clicks the "Approve posting to Department C" button. The user's selection is sent to the server, and the final posting destination is confirmed.

[1812] Examples of prompt statements

[1813] If the sender is "Company A", the delivery item is "Type B Parts" and the size is "Medium", please suggest an appropriate drop-off location based on past data.

[1814] In this way, users can efficiently sort mail and parcels that are not addressed to their destination. By utilizing generative AI models and a database management system, the system can accurately suggest possible recipients based on past data.

[1815] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1816] Step 1:

[1817] Data Entry

[1818] The user uses the terminal to enter data such as sender information and the type, size and format of the delivery item.

[1819] Input: sender information (e.g. Company A), type of delivery (e.g. Type B parts), size (e.g. medium size)

[1820] Specific operation: The user enters "Sender: Company A, Delivery: Type B parts, Size: Medium" into the input form on the terminal and clicks the send button.

[1821] Output: The entered data is sent to the server.

[1822] Step 2:

[1823] Database Reference

[1824] The server uses a database management system (e.g. MySQL) to look up details of past mailings and deliveries.

[1825] Input: Data entered by the user (sender information, type of delivery, size)

[1826] Specific operation: The server executes an SQL query and sends the command "SELECT FROM mailing history WHERE sender = 'Company A' AND delivery = 'Parts of type B'" to the database.

[1827] Output: Detailed information about past mailing destinations and deliveries is obtained.

[1828] Step 3:

[1829] Generating recipient candidates

[1830] The server uses a generative AI model (e.g. GPT-4) to generate potential recipients based on the input data and past data.

[1831] Input: Data entered by the user, details of past mailings and deliveries retrieved from a database

[1832] Specific operation: The server sends a prompt to the generative AI model: "If the sender is Company A and the delivery is a type B part, please suggest an appropriate delivery destination based on past data." The generative AI model responds: "Department C is appropriate."

[1833] Output: A list of possible recipients (e.g. Department C) is generated.

[1834] Step 4:

[1835] Suggestions for possible recipients

[1836] The server transmits the generated candidate mailbox destinations to the terminal and suggests them to the user.

[1837] Input: Generated recipient candidates (e.g. Department C)

[1838] Specific operation: The server sends a message to the terminal saying, "We suggest posting to Department C." The content of the suggestion is displayed on the terminal screen.

[1839] Output: The suggestions are displayed to the user.

[1840] Step 5:

[1841] User confirmation and final decision

[1842] The user checks the proposed destination candidates and decides on the final destination.

[1843] Input: Suggested recipient (e.g. Department C)

[1844] Specific operation: The user checks the proposal displayed on the terminal screen and clicks the "Approve submission to Department C" button. The user's selection is sent to the server.

[1845] Output: The final posting destination is confirmed.

[1846] In this way, users can efficiently sort mail and parcels that are not addressed to their destination. By utilizing generative AI models and a database management system, the system can accurately suggest possible recipients based on past data.

[1847] (Application example 2)

[1848] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal".

[1849] The conventional method of sorting mail and parcels with unknown addresses requires the worker to check the work experience and sorting list each time, which is inefficient and prone to human error. In addition, there is no system that uses past data to suggest the optimal destination, so there is a need to improve the accuracy and efficiency of the sorting process.

[1850] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1851] In this invention, the server includes a means for suggesting possible post destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible post destinations based on past post destinations and detailed information on the delivery, a means installed in the smartphone for suggesting the optimal post destination based on past data, and a means for saving the history of the suggested post destinations and improving the accuracy of the suggestions from the next time onwards. This makes it possible to significantly improve the efficiency and accuracy of the sorting work for mail and parcels with unknown addresses.

[1852] "Unknown address mail or parcel" refers to mail or parcel that cannot be delivered to a specific address through normal delivery procedures because the address specified by the sender is unclear.

[1853] "Sorting work" refers to the task of classifying mail and parcels based on their destination and content, and then distributing them to the appropriate destination.

[1854] "Sender" refers to the individual or organization sending the mail or parcel.

[1855] "Type of delivery" refers to the category of the contents of a mail or parcel, such as documents, merchandise, parts, etc.

[1856] "Size" refers to the physical size of a mail item or parcel, and is classified as small, medium, large, etc.

[1857] "Format" refers to the shape and type of packaging of a mail or parcel, such as an envelope, box, or pallet.

[1858] "Potential drop-off locations" refer to locations or departments to which undeliverable mail or parcels may be delivered.

[1859] "Previous mailing destination" refers to a location or department to which undeliverable mail or parcels have previously been delivered.

[1860] "Detailed information" refers to specific data and history about mail or parcels.

[1861] A "smartphone" is a type of mobile phone and a multi-function device that can connect to the Internet and install applications.

[1862] "Saving history" means recording information about past mailing destinations and deliveries in a database so that it can be referenced later.

[1863] "Suggestion accuracy" refers to the system's ability to accurately suggest the optimal drop-off destination.

[1864] As an embodiment of the present invention, a system for streamlining the sorting of mail and parcels with unknown destinations is constructed. A specific embodiment of the system is described below.

[1865] System Configuration

[1866] Hardware:

[1867] Smartphone (iOS or Android)

[1868] software:

[1869] Python

[1870] Pandas (Data Processing)

[1871] Scikit-learn (machine learning model)

[1872] Program processing explanation

[1873] Data Loading:

[1874] The server reads past delivery data from a CSV file, which gives information about the sender, type of delivery, size, format, and past delivery destination.

[1875] Split features and labels:

[1876] Based on the data it reads, the server classifies the sender, type of delivery, size, and format as features, and the delivery destination as a label.

[1877] Model training:

[1878] The server trains a machine learning model using K-nearest neighbors (KNN) to improve its ability to suggest optimal drop destinations based on past data.

[1879] Suggested destinations:

[1880] When a user uses their smartphone to input information about a new delivery (sender, type of delivery, size, format), the server uses the trained model to suggest the optimal delivery location.

[1881] Save history:

[1882] The server stores the history of suggested destinations in a database, which improves the accuracy of suggestions from the next time onwards.

[1883] Examples

[1884] For example, if the sender is "Company A," the delivery item is a "Type B part," the size is "medium," and the format is "box," it will suggest posting it to "Department C" based on past data.

[1885] Examples of prompt statements

[1886] Please use the following information to suggest the best place to send your message.

[1887] From: Company A

[1888] Delivery type: B-type parts

[1889] Size: Medium

[1890] Format: Box

[1891] In this way, a system can be realized that improves the efficiency and accuracy of sorting mail and parcels with unknown destinations.

[1892] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1893] Step 1:

[1894] Data Loading

[1895] The server reads past delivery data from a CSV file. The path of the CSV file is specified as input. The server reads the CSV file in data frame format using the Pandas library and obtains information on the sender, type of delivery, size, format, and past recipients. A data frame is generated as output.

[1896] Step 2:

[1897] Splitting features and labels

[1898] Based on the data read, the server separates the sender, type, size, and format of the delivery as features, and the recipient as a label. The data frame generated in step 1 is used as input. The server splits the features and labels into separate data frames, in preparation for training the machine learning model. As output, a feature data frame and a label data frame are generated.

[1899] Step 3:

[1900] Model Training

[1901] The server trains a machine learning model using K-nearest neighbors (KNN). As input, the feature and label dataframes generated in step 2 are used. The server trains the KNN model using the Scikit-learn library to improve its ability to suggest optimal drop destinations based on past data. As output, a trained KNN model is generated.

[1902] Step 4:

[1903] Suggested destination

[1904] A user inputs new parcel information (sender, parcel type, size, format) using a smartphone. The parcel information entered by the user is used as input. The server uses the trained KNN model to suggest the optimal delivery destination based on the input information. As output, the suggested delivery destination is displayed to the user.

[1905] Step 5:

[1906] History storage

[1907] The server saves the history of suggested destinations in a database. As input, the proposed destinations in step 4 and the delivery information entered by the user are used. The server records this information in the database to improve the accuracy of future suggestions. As output, the history data is saved in the database.

[1908] In this way, a system can be realized that improves the efficiency and accuracy of sorting mail and parcels with unknown destinations.

[1909] In addition, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1910] "Example 1"

[1911] As an embodiment of the present invention, a system incorporating an emotion engine is provided. The system recognizes the user's emotion and adjusts the candidate destinations based on the emotion. Specifically, if the system recognizes that the user is stressed, the system reduces the number of candidate destinations and simplifies the options. Also, if the system recognizes that the user is relaxed, the system provides more candidate destinations, giving the user more freedom of choice. Furthermore, the emotion engine also adjusts the way in which the candidate destinations are presented based on the user's emotion. For example, if the system recognizes that the user is anxious, the system does not display a list of candidate destinations, but directly presents the most likely destination. This allows the user to make a decision quickly.

[1912] "Example 2"

[1913] As an embodiment of the present invention, a system incorporating an emotion engine is provided. The system recognizes the user's emotion and adjusts the candidate destinations based on the emotion. Specifically, if the system recognizes that the user is stressed, the system reduces the number of candidate destinations and simplifies the options. Also, if the system recognizes that the user is relaxed, the system provides more candidate destinations, giving the user more freedom of choice. Furthermore, the emotion engine also adjusts the way in which the candidate destinations are presented based on the user's emotion. For example, if the system recognizes that the user is anxious, the system does not display a list of candidate destinations, but directly presents the most likely destination. This allows the user to make a decision quickly.

[1914] The process flow of each embodiment will be described below.

[1915] "Example 1"

[1916] Step 1: The system recognizes the user's emotions using an emotion engine, which infers emotions from the user's facial expressions, tone of voice, choice of words, etc.

[1917] Step 2: After the emotion engine recognizes the user's emotion, the system will adjust the suggested destinations based on the emotion. For example, if the system recognizes that the user is stressed, it will reduce the number of suggested destinations and simplify the options.

[1918] Step 3: On the other hand, if the system detects that the user is relaxed, it will provide more possible destinations, giving the user more freedom of choice.

[1919] Step 4: Furthermore, the emotion engine will also adjust the way in which the system presents the possible destinations based on the user's emotions. For example, if the system detects that the user is in a hurry, it will directly present the most likely destination instead of displaying a list of possible destinations, allowing the user to make a decision quickly.

[1920] Example 1

[1921] Next, a description will be given of Example 1 of the embodiment 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal".

[1922] Conventional methods for sorting mail and parcels with unknown addresses are inefficient because they require the person in charge to check the work experience and sorting list each time. In addition, there is a problem that appropriate support cannot be provided to users who feel stressed or impatient because the system cannot respond flexibly to the user's emotional state. This has led to a demand for more efficient sorting work and improved user satisfaction.

[1923] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1924] In this invention, the server includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on the emotions. This makes it possible to streamline the sorting of mail and parcels with unknown addresses and to respond flexibly to the user's emotional state.

[1925] "Sender" refers to the individual or organization sending the mail or parcel.

[1926] "Delivery" refers to goods or documents sent by mail or courier.

[1927] "Type" is an attribute that indicates the classification of the delivery, and includes, for example, documents, parts, and merchandise.

[1928] "Size" is an attribute that indicates the physical size of a delivery item, and includes, for example, length, width, height, weight, and the like.

[1929] "Form" is an attribute that indicates the shape of the delivery item or the state of packaging, and includes, for example, a box, an envelope, a pallet, and the like.

[1930] "Potential drop-off locations" refers to locations or departments where deliveries may be dropped off.

[1931] "Previous mailbox locations" refers to records of locations or departments where deliveries have been previously mailed.

[1932] "Detailed information" refers to specific data and history about the delivery.

[1933] "Emotion recognition" refers to identifying a user's emotional state by analyzing their facial expressions and tone of voice.

[1934] "Adjust" refers to changing the system's behavior or suggestions to suit specific conditions or circumstances.

[1935] The present invention is a system for streamlining the sorting of mail and parcels with unknown destinations. This system includes a means for suggesting possible destinations based on the sender and the type, size, and format of the delivery, a means for suggesting possible destinations based on past destinations and detailed information on the delivery, and a means for recognizing the user's emotions and adjusting the possible destinations based on those emotions.

[1936] Hardware and Software Used

[1937] Hardware

[1938] Server: Analyzes and stores data.

[1939] Terminal: Provides a user interface and is equipped with a camera and microphone for recognizing the user's emotions.

[1940] Scanner: Scan postal and courier labels.

[1941] software

[1942] Google Cloud Vision API: Converts paper data into digital data using OCR (optical character recognition) technology.

[1943] MySQL: A database management system that stores analysis results and past submission data.

[1944] Microsoft Azure Emotion API: Recognizes emotions by analyzing the user's facial expressions and tone of voice.

[1945] System Operation

[1946] Data Entry and Analysis

[1947] The server receives data such as sender information for mail and parcels, and the type, size, and format of the delivery. Specifically, it scans the mail and parcel labels and converts them from paper to digital data using Google Cloud Vision API. The server analyzes the received data and identifies the sender and the type, size, and format of the delivery. The analysis results are stored in a MySQL database.

[1948] Suggestions for possible recipients

[1949] The server searches a database of past delivery destinations based on the analysis results and suggests the most appropriate delivery destination. For example, if the sender is "Company A" and the delivery item is "Type B parts," it suggests delivery to "Department C" based on past data. Also, if the size of the delivery item falls within a certain range, it suggests delivery to a specific location.

[1950] Emotion engine adjustment

[1951] The device uses a camera and microphone to analyze the user's facial expressions and tone of voice to recognize the user's emotions. It uses Microsoft Azure's Emotion API. If the server recognizes that the user is stressed, it reduces the number of possible delivery candidates and simplifies the options. Conversely, if the server recognizes that the user is relaxed, it provides more possible delivery candidates.

[1952] Suggestion of possible recipients

[1953] The server sends the adjusted drop-off candidates to the terminal and presents them to the user. If the server detects that the user is in a hurry, it will directly present the most likely drop-off destination, allowing the user to make a quick decision.

[1954] Examples of concrete examples and prompts

[1955] Examples

[1956] When a user receives a "Part B" from "Company A", the system will suggest dropping it off at "Division C" based on past data. If the system detects that the user is stressed, it will only suggest dropping it off at "Division C".

[1957] Examples of prompt statements

[1958] "We have received part B from company A. Please suggest the best delivery destination based on past data. Also, please adjust the suggestions based on user sentiment."

[1959] The above is an embodiment of the present invention.

[1960] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1961] Step 1:

[1962] Data Entry

[1963] The server receives data such as sender information for mail or parcels, and the type, size, and format of the delivery. As input, it scans the label of the mail or parcel to obtain image data. In concrete terms, the server obtains image data from the scanner and calls the Google Cloud Vision API to extract text information from the image data. As output, it obtains the extracted text information.

[1964] Step 2:

[1965] 4. Data Analysis

[1966] The server analyzes the text information obtained in step 1 to identify the sender, type of delivery, size, and format. As input, it receives the extracted text information. In concrete terms, the server analyzes the text information to identify the sender, type of delivery, size, and format. As output, it obtains the analysis results, which are stored in a MySQL database.

[1967] Step 3:

[1968] Suggestions for possible recipients

[1969] The server searches the database of past destinations based on the analysis results obtained in step 2, and proposes the most suitable destination candidates. It receives the analysis results as input. In concrete terms, the server queries the MySQL database to retrieve the past destination data. As output, it obtains a list of optimal destination candidates.

[1970] Step 4:

[1971] Emotion engine adjustment

[1972] To recognize the user's emotions, the device uses a camera and microphone to analyze the user's facial expressions and tone of voice. It receives the user's facial expression and voice data as input. Specifically, the device calls Microsoft Azure's Emotion API to analyze the user's emotions. The output is the user's emotional state. The server adjusts the number of possible recipients based on the user's emotional state.

[1973] Step 5:

[1974] Suggestion of possible recipients

[1975] The server sends the adjusted list of destination candidates in step 4 to the terminal and presents them to the user. As input, it receives the adjusted list of destination candidates. In concrete terms, the server sends the adjusted destination candidates to the terminal, and the terminal displays the destination candidates on the display. As output, the destination candidates presented to the user are obtained.

[1976] (Application example 1)

[1977] Next, a description will be given of Application Example 1 of Form Example 1...

Claims

1. a suggestion means for generating one or more candidate destinations for the delivery based on a sender's name, the sender's address, the type of delivery from the sender, the size of the delivery, and the format of the delivery by inputting a specific prompt sentence into a generation AI model, and for suggesting the generated one or more candidate destinations to a user who is a sorter; an emotion engine that recognizes the emotion of the user who has been suggested a delivery destination candidate of the delivery; an adjustment means for adjusting the number of candidate recipients based on the emotion; A system including:

2. The system according to claim 1 , wherein the adjustment means generates the most likely destination candidate as the destination candidate when the user is feeling stressed, and generates multiple destination candidates when the user is relaxed.

3. The system of claim 1 , wherein the suggestion means generates one or more candidate destinations for the delivery, including destinations of past deliveries by the sender.

Images