Shopping support system, sales data processing apparatus, and shopping support program

Abstract

To propose a new production which is not made with an existing kit or based on an existing recipe, to a user.SOLUTION: A shopping support system includes: first storage means which stores commodity information which is information on a commodity that a store sells; second storage means which stores transaction information which is the commodity information registered as a transaction object; receiving means which receives an input of request from a customer; generation means which generates a new production which can be made using at least part of the transaction information, based on learning result data which is generated by machine learning that uses information on existing productions as training data; and presentation means which presents information output by the generation means so as to be recognized by the customer.SELECTED DRAWING: Figure 13

Description

Technical Field

[0001] Embodiments of the present invention relate to a shopping support system, a sales data processing device, and a shopping support program.

Background Art

[0002] In a conventional system, it is possible to provide information extracted from the existing method of making a product to a customer who visits a store to purchase materials and tools for making the product. However, if proposals are made to the user only based on existing kits and methods, the proposals will be limited to the existing information.

[0003] Regarding a product or its arrangement, when consulting an expert such as a human store clerk, it is possible to obtain information with creativity that is not existing information. On the other hand, since the proposals possible with the conventional system are limited to the existing information as described above, there is a disadvantage that it is impossible to propose a new creative product to the user or a product that stimulates creativity.

[0004] Recommendations using artificial intelligence such as Patent Document 1 have been proposed, but the technique disclosed in Patent Document 1 cannot solve the above-mentioned disadvantages.

Summary of the Invention

Problems to be Solved by the Invention

[0005] The problem to be solved by the present invention is to provide a shopping support system, a sales data processing device, and a shopping support program that can propose a new product that does not depend on existing kits or existing methods to the user.

Means for Solving the Problems

[0006] The shopping support system according to the embodiment includes: a first storage means for storing product information which is information on products handled by a store; a second storage means for storing transaction information which is the product information registered as an object of a transaction; a reception means for receiving an input of a customer's request; a generation means for generating a new product that can be produced using at least a part of the transaction information based on learning result data generated by machine learning with information on an existing product as teacher data; and a presentation means for presenting the information output by the generation means in a recognizable manner to the customer.

Brief Description of Drawings

[0007]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

DETAILED DESCRIPTION OF THE INVENTION

[0008] (First Embodiment) The embodiments will be described with reference to the drawings. FIG. 1 is a diagram showing an example of the configuration of the shopping support system according to the present embodiment. The shopping support system according to the present embodiment (hereinafter simply referred to as the system) includes an application server 1, an artificial intelligence server 3, and a mobile POS 5. Here, "POS" is an abbreviation of "Point Of Sale" and means "sales point information management". The mobile POS 5 is an embodiment of a sales data processing device that processes registration of products and settlement of registered products.

[0009] The mobile POS 5 is a mobile terminal (for example, a portable terminal device such as a smartphone or a tablet-type terminal) on which a POS app (abbreviation of application software) operates.

[0010] The mobile POS 5 may be realized by a terminal device owned by a customer (user), or may be lent by a store to a customer during shopping. Further, the mobile POS 5, which is a terminal device lent by the store, may be attached to a shopping cart and used. Here, the shopping cart is a handcart or a trolley for carrying products, and includes, for example, a basket-shaped container on which products are placed or stored, a frame that supports the container, and wheels attached to the lower part of the frame.

[0011] The POS app is provided so as to be downloadable, for example, by the application server 1. The mobile POS 5 provides a POS function by starting the installed POS app. The mobile POS 5 performs a process of registering product information by an operation of a customer during shopping.

[0012] The application server 1 can communicate (send and receive information) with the mobile POS 5 and the artificial intelligence server 3 via the network 2. The network 2 is a public network such as the Internet, for example. The application server 1 collaborates with the POS application to perform processing according to the operations received by the mobile POS 5 and communicate with the mobile POS 5.

[0013] The artificial intelligence server 3 collaborates with the application server 1 to output information (response) in accordance with the requests received by the POS application. Note that the artificial intelligence server 3 performs supervised learning by artificial intelligence (AI) and generates the above-mentioned response based on the results of the learning.

[0014] The system of this embodiment proposes new products that are not based on existing kits or manufacturing methods to users. A kit is a set of articles for obtaining a product (such as a chair, desk, or shelf), which is a collection of materials, parts, tools, and instructions (such as design drawings, material lists, and assembly instructions) of a predetermined size. Kits are products sold in stores such as home centers. Also, the manufacturing methods of existing products can be obtained via a network such as the Internet or sold as publications. If only existing kits and manufacturing methods are proposed to users, new products cannot be proposed. Therefore, in this embodiment, existing kits and manufacturing methods are learned by the artificial intelligence server 3, and the learned artificial intelligence server 3 is used to obtain (generate) proposals for new products that do not exist. When presenting the proposals, information such as the manufacturing method, materials, and recommended tools for using the new products is also presented together.

[0015] Figure 2 is a block diagram showing an example of the configuration of the application server 1. The application server 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a communication unit 14, a storage unit 19, and the like.

[0016] The CPU 11 is an example of a processor and comprehensively controls each part of the application server 1. The ROM 12 stores various programs. The RAM 13 is a workspace for expanding programs and various data. The CPU 11, ROM 12, and RAM 13 are connected via a bus or the like and constitute a control unit 10 of a computer configuration.

[0017] The communication unit 14 is a communication interface that communicably connects the control unit 10 to other devices (such as the mobile POS 5 and the artificial intelligence server 3) via the network 2.

[0018] The storage unit 19 has a storage medium such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or a flash memory, and maintains the stored content even when the power is turned off. The storage unit 19 stores programs 191 executable by the CPU 11, various setting information such as a product master 192 and a purchase history 193. Further, the storage unit 19 stores transaction information 195 received from each store. The storage unit 19 is an example of a first storage means for storing product information, which is information on products handled by the store, and is an example of a second storage means for storing transaction information, which is information on products registered as transaction targets.

[0019] FIG. 3 is a diagram showing an example of items included in the product master 192. The product master 192 is an example of product information, and information on products handled by the store (product information) is summarized in, for example, a table format. The items of the product master 192 are, for example, as follows. · Product code · Product name · Unit price · Image · Verification data · Display location

[0020] The product code is an example of information (identification information) that can identify a product, and is, for example, a JAN code. Other information (such as product name and unit price) is stored in association with the product code.

[0021] What is recorded in the "Product Name" field is the name of the product. What is recorded in the "Unit Price" field is the price of one product.

[0022] What is recorded in the "Image" field is an image of the appearance of the product. What is recorded in the "Verification Data" field is the reference feature amount. The device related to the sales data processing compares the feature amount of the image (captured image) captured and output by a camera or the like with the verification data, and recognizes the product by obtaining the product code associated with the matching verification data.

[0023] What is recorded in the "Display Location" field is information indicating the location where the product is displayed (display location), for example, the name of the display shelf or sales floor, the floor number, etc.

[0024] When registering a product, the mobile POS 5 obtains product information by referring to the product master 192 stored in the application server 1.

[0025] Figure 4 is a diagram showing an example of transaction information 195. The transaction information 195 is information on the product that is the subject of the transaction (that is, the information on the product registered as being purchased by the customer), and is summarized in a table format, for example. The items of the transaction information 195 are as follows, for example. · Transaction ID · Date and Time · Store ID · Customer Information · Product Information · Amount

[0026] The transaction ID is an example of information (identification information) for identifying a transaction. The transaction ID is automatically numbered (assigned), for example, at the time of the first product registration process related to the transaction. The date and time is the date and time when the transaction of the record was made. The store ID is the identification information of the store where the transaction of the record was made. The customer information is information (such as customer ID, name, contact information, etc.) that can identify (or specify) the customer. The amount is the total (aggregate) price of all products purchased in the transaction of the record.

[0027] Here, for a combination of a transaction ID, a date and time, and a store ID, although there is one amount, multiple pieces of product information may be associated. The product information includes, for example, the following items. · Product code · Product name · Quantity · Price

[0028] The product code is as described above. The quantity is the quantity (number, weight, volume, etc.) of the product indicated by the product code purchased in the transaction of the record. The price is the value obtained by multiplying the unit price (or selling price) by the quantity.

[0029] Each time a process of registering product information is executed in the mobile POS 5, the application server 1 causes a new record to be stored (registered) in the transaction information 195. Also, when a process of deleting a product registered in the mobile POS 5 is executed, the application server 1 deletes the record in the transaction information 195.

[0030] Furthermore, when a transaction in the mobile POS 5 is completed, the application server 1 stores (accumulates) and manages the information (transaction information 195) of the completed transaction as a purchase history 193 in the storage unit 19.

[0031] FIG. 5 is a diagram showing an example of the purchase history 193. The purchase history 193 is a summary of information on products purchased by a customer in the past, for example, in the form of a table, and includes, for example, the same items as the transaction information 195.

[0032] In response to an inquiry (request) from the mobile POS 5, the application server 1 extracts a record that matches the specified customer information (for example, a customer ID) from the purchase history 193.

[0033] Note that the program 191, product master 192, purchase history 193, and transaction information 195 do not necessarily have to be stored in the same storage unit 19. Also, the purchase history 193 may be obtained from an external system such as an electronic receipt server. The electronic receipt server accumulates and manages the electronic receipts related to the contracted stores. The electronic receipt is the electronic data conversion of the receipt (which displays the details of the transaction) that was conventionally handed to the customer as a printed matter.

[0034] FIG. 6 is a block diagram showing an example of the configuration of the artificial intelligence server 3. The artificial intelligence server 3 includes a CPU 31, a ROM 32, a RAM 33, a communication unit 34, a storage unit 39, etc.

[0035] The CPU 31 is an example of a processor and comprehensively controls each part of the artificial intelligence server 3. The ROM 32 stores various programs. The RAM 33 is a workspace for developing programs and various data. The CPU 31, ROM 32, and RAM 33 are connected via a bus or the like and constitute a control unit 30 of a computer configuration.

[0036] The communication unit 34 is a communication interface that communicably connects the control unit 30 and other devices (such as the application server 1, etc.) via the network 2. Also, the communication unit 34 communicably connects the control unit 30 and other devices (such as the application server 1, etc.) via a network such as a LAN (Local Area Network) installed in the store.

[0037] The storage unit 39 has a storage medium such as an HDD, SSD, or flash memory, and maintains the stored content even when the power is turned off. The storage unit 39 stores the program 391 executable by the CPU 31, the learning result data 400, etc. The storage unit 39 stores the learning result data 400 so that it can be referred to from the application server 1.

[0038] The CPU 31 of the artificial intelligence server 3 performs supervised learning by operating according to the program 391. In supervised learning, the artificial intelligence server 3, for example, takes in the teacher data 200, uses the teacher data 200 to cause a generative AI (for example, a learning model such as a large language model) to execute learning by a known method, and obtains (generates) the learning result data 400 that is the result of the learning.

[0039] The generative AI is, for example, a known Transformer model including an encoder layer and a decoder layer. The generative AI is configured as a program 391 executable by the CPU 31 of the artificial intelligence server 3. Note that the program 391 and the learning result data 400 may be downloaded and provided to the application server 1 or the mobile POS 5 via the network 2, or may be stored in a non-transitory recording medium such as a CD-ROM and then provided to the application server 1 or the mobile POS 5.

[0040] The teacher data 200 is, for example, the off-the-shelf information 201, the knowledge information 202, and the purchase history 193 (described above). From the purchase history 193, information for estimating combinations of products and tools can be obtained.

[0041] The off-the-shelf information 201 is an example of information regarding existing manufactured goods and is off-the-shelf manufactured goods information. The manufactured goods are goods that a customer manufactures and are not goods sold as finished products in a store.

[0042] Specifically, the off-the-shelf information 201 is, for example, information regarding kits sold by a home center. A kit is a set of articles for obtaining a manufactured good (for example, a chair, a table, a shelf), and includes materials, parts, tools, and an assembly instruction manual (for example, a design drawing, a material list, an assembly description text) of a predetermined size. When the materials and parts included in the kit are assembled according to the instruction manual, the manufactured good is completed. For example, the above description is suitable as the off-the-shelf information 201.

[0043] In addition, the ready-made information 201 is information that explains how to make a product and can be obtained via the network 2, for example. This information includes information such as design drawings of the product, materials, examples of suitable materials and products, dimensions of each component, tools recommended for use, and assembly instructions.

[0044] The knowledge information 202 is information about products recorded independently by, for example, the clerks at a home center. This information includes information about products (kits, materials, tools) and their associations.

[0045] Note that based on the purchase history 193, the artificial intelligence server 3 can learn combinations of purchased products (kits, materials, tools), etc.

[0046] The learning result data 400 is an example of data generated by machine learning using information about existing products as teacher data. The learning result data 400 obtained by learning using the teacher data 200 is, for example, product information 401, tool information 402, and additional information 403. Note that the learning result data 400 shown here is merely one specific example and is not limited to this in implementation. That is, the learning result data 400 may be other than the product information 401, tool information 402, and additional information 403.

[0047] The learning result data 400 is used for recommending new products and products to customers, presenting materials and tools, and guiding customers. More specifically, when the application server 1 recommends new products and products to customers, presents materials and tools, and guides customers via the mobile POS 5, it makes an inquiry to the artificial intelligence server 3 and receives information provision. At that time, the artificial intelligence server 3 responds to the inquiry from the application server 1 using the learning result data 400. Here, the new product presented is a product that can be made using at least a part of the product information and the transaction information 195.

[0048] FIG. 7 is a diagram showing an example of the workpiece information 401. Here, the workpiece information 401 is shown in the form of a table for convenience, but in practice, the learning result data 400 is not limited to the table form.

[0049] The workpiece information 401 includes information indicating, for example, the name, category, materials (material 1, material 2,..., material N), cost, working time, additional conditions, etc. of the workpiece. These pieces of information are associated with, for example, a unique code (ID) and stored in the storage unit 39.

[0050] The category is information indicating the classification of the workpiece, such as a chair, a desk, a shelf, etc.

[0051] The information indicating the cost shows the cost (expenses) required to manufacture the workpiece, and may be information indicating a degree such as large, medium, small, etc., or may also be a specific numerical value.

[0052] The information indicating the working time is information regarding the time required to complete the workpiece, and may be information indicating a degree such as large, medium, small, etc., or may also be a specific numerical value.

[0053] The additional conditions are additional information such as information differentiating the same or similar workpieces, such as "inexpensive", "luxurious", "simple". "Inexpensive" indicates that the cost required for manufacturing is low compared to the same category ("cost" is "small"). "Luxurious" indicates that the appearance of the workpiece is expected to have a luxurious finish rather than a cheap look compared to the same category. "Simple" indicates that the technology required for manufacturing is not high (it is possible to manufacture even with low technology) or that the time required for manufacturing is short ("working time" is "small") compared to the same category.

[0054] FIG. 8 is a diagram showing an example of the tool information 402. Here, the tool information 402 is shown in the form of a table for convenience, but in practice, the learning result data 400 is not limited to the table form.

[0055] The tool information 402 is information that stores, for example, information about tools (such as tools) used in the production of a manufactured item, associated with the same ID as the manufactured item information 401. In the tool information 402, information indicating, for example, the material of the manufactured item may be further stored in association with the information indicating the tool.

[0056] FIG. 9 is a diagram showing an example of the additional information 403. Here, the additional information 403 is shown in the form of a table for convenience, but in practice, the learning result data 400 is not limited to the table form. The items included in the additional information 403 are, for example, material, additional conditions, and appearance probability.

[0057] The additional information 403 is information that stores information indicating additional conditions in association with information indicating the material. The additional conditions are, as described above, words such as "inexpensive", "luxury", "simple", and are additional information such as information that differentiates the same type or similar manufactured items. The additional information 403 indicates that by using the corresponding material, the additional conditions can be satisfied or the possibility of satisfying them is increased.

[0058] The appearance probability is the appearance probability of each word, and is generated by the above-described generation AI (realized by the CPU 31 of the artificial intelligence server 3 executing the program 391).

[0059] More specifically, during the learning of the generation AI, example question sentences are input to the encoder layer of the transformer model, and example answer sentences are input to the decoder layer. Here, both the example question sentences and the example answer sentences are the teacher data 200. The example question sentences are sentences assuming questions from customers, and the example answer sentences are sentences assuming answers to customers. Through the above learning, the generation AI generates the appearance probability of each word used when generating an answer sentence for an actual customer's question sentence. Note that the question sentence may be a text (question sentence) obtained by converting the voice of the question input by the customer by the generation AI. Similarly, the answer sentence may be converted from the text (answer sentence) to the pseudo voice of the artificial intelligence server 3 by the generation AI.

[0060] Also, during the inference of the generative AI, when a question sentence example from a customer is input to the encoder layer of the Transformer model, an answer sentence that proposes a new product or the like is output from the decoder layer. Further, when an answer sentence that proposes a new product or the like is input to the decoder layer, the appearance probability of each word in the answer sentence changes. That is, the appearance probability of each word in the answer sentence is updated.

[0061] FIG. 10 is a block diagram showing an example of the configuration of the mobile POS 5. The mobile POS 5 includes a CPU 51, a ROM 52, a RAM 53, a communication unit 54, a display unit 55, an operation input unit 56, an audio output unit 57, an imaging unit 58, a storage unit 59, and the like.

[0062] The CPU 51 is an example of a processor and comprehensively controls each part of the mobile POS 5. The ROM 52 stores various programs. The RAM 53 is a workspace for expanding programs and various data. The CPU 51, the ROM 52, and the RAM 53 are connected via a bus or the like and constitute a control unit 50 of a computer configuration.

[0063] The communication unit 54 is a communication interface that communicably connects the control unit 50 and another device (such as the application server 1) via the network 2.

[0064] The display unit 55 is an example of a presentation means that presents the information output by the generation means in a recognizable manner to the customer. The display unit 55 has a display device such as an LCD (Liquid Crystal Display) and displays various information under the control of the CPU 51.

[0065] The operation input unit 56 is an example of a reception means that receives the input of a customer's request and performs information exchange (transmission and reception) in an interactive format using voice or characters.

[0066] The operation input unit 56 has input devices such as a touch panel and a microphone provided on the surface of the display unit 55, and outputs the operation content input via the input device to the CPU 51. The touch panel accepts operations according to the display content of the display unit 55. For example, if the display unit 55 is displaying a software keyboard, the touch panel accepts an operation for inputting characters. The microphone also accepts voice as input and outputs it to the CPU 51. The CPU 51 recognizes the input voice as words according to the process being executed at the time of voice input.

[0067] The voice output unit 57 has a speaker and a buzzer, and outputs voice under the control of the CPU 51.

[0068] The imaging unit 58 is, for example, a camera provided in the terminal device. The mobile POS 5 has a function of decoding a code symbol (for example, a barcode or a two-dimensional code) included in an image (captured image) captured and output by the imaging unit 58. If the code symbol encodes, for example, a product code, this function outputs the product code.

[0069] Here, the operation of the user (customer) of the mobile POS 5 capturing a code symbol attached to a product with the imaging unit 58 of the mobile POS 5 and having the mobile POS 5 read it is sometimes referred to as "scanning" below.

[0070] Note that the mobile POS 5 may also acquire a value obtained by the above device reading and decoding a code symbol by receiving the output from a device such as a wired or wirelessly connected code scanner.

[0071] The storage unit 59 has a storage medium such as an HDD, an SSD, or a flash memory, and maintains the stored content even when the power is turned off. The storage unit 59 stores programs 591 executable by the CPU 51, transaction information, and the like.

[0072] The program 591 is application software for making the mobile POS 5 function.

[0073] The transaction information is information on products registered as being purchased by the customer (i.e., products subject to the transaction), summarized in, for example, a table format. Since the items of the transaction information are substantially the same as those of the transaction information 195, the description thereof is omitted. Each time a product is registered, the mobile POS 5 acquires product information from the application server 1.

[0074] Note that the transaction information may be stored in the RAM 53 instead of the storage unit 59. The storage unit 59 or the RAM 53 stores the transaction information, which is information on the registered products (registered items), from the start to the completion of at least one transaction. The transaction is completed when the accounting process is completed. The accounting process is a process of receiving payment for the registered products from the customer.

[0075] FIG. 11 is a block diagram showing an example of various functional units provided in each of the control units 10, 30, 50 and their cooperation. The control units 10, 30, 50 provide the various functional units shown in FIG. 11 by executing the programs 191, 391, 591, respectively.

[0076] The control unit 50 of the mobile POS 5 functions as a registration processing unit 501, an interactive processing unit 502, and an accounting processing unit 503. The control unit 10 of the application server 1 functions as a registration processing unit 101, an AI cooperation unit 102, and an accounting processing unit 103. The control unit 30 of the artificial intelligence server 3 functions as a learning processing unit 301, a cooperation unit 302, and an update processing unit 303. Here, the control unit 30 is an example of a generation means, and generates information for manufacturing a new product by inference based on the learning result data 400.

[0077] The learning processing unit 301 obtains the learning result data 400 by causing a learning model to perform machine learning using at least one of the ready-made information 201 indicating the materials, tools, and manufacturing methods of existing products, the knowledge information 202 of the store employees working in the store, and the customer purchase history 193 as teacher data.

[0078] The learning processing unit 301 is a functional unit corresponding to the above-described generative AI. By taking in the teacher data 200 and performing learning, it obtains (generates) the learning result data 400. The teacher data 200 may be obtained from the application server 1 or may be provided from other external systems.

[0079] Further, the learning processing unit 301 causes a learning model (transformer) to learn question sentence examples and answer sentence examples generated from at least one of the instructional data of the materials, tools, and manufacturing methods of existing products, the knowledge information 202 of store employees working in the store, and the purchase history 193 of customers, thereby obtaining the appearance probability of each word used when generating an answer sentence for an actual customer's question sentence.

[0080] The registration processing unit 501 collaborates with the registration processing unit 101 to perform processing for registering product information.

[0081] The dialogue processing unit 502, the AI cooperation unit 102, and the cooperation unit 302 collaborate to engage in consultations regarding products by customers and return recommendations.

[0082] The dialogue processing unit 502 performs dialogue processing (chat) with the user (customer). The dialogue processing is the transmission and reception of information in a dialogue format, which is a process of understanding each other's statements, negotiating, and obtaining a conclusion. The AI cooperation unit 102, in collaboration with the cooperation unit 302, generates a proposal (or answer) for the user's question received from the dialogue processing unit 502 and transmits the generated proposal to the mobile POS 5.

[0083] The accounting processing unit 503 collaborates with the accounting processing unit 103 to perform processing related to accounting (payment of money by the customer, settlement).

[0084] After the accounting process, the update processing unit 303 updates the learning result data 400 with the accounted transaction information 195 obtained from the accounting processing unit 103. Here, the transaction information 195 obtained from the accounting processing unit 103 is the feedback information of the consultation and recommendation if it is after the consultation and recommendation of the product.

[0085] FIG. 12 is a sequence diagram showing the flow of processes executed by each of the control units 10, 30, and 50. Note that the user shown in FIG. 12 is the user of the mobile POS 5, that is, the customer during shopping. The customer during shopping (user) scans the code symbol attached to the product and puts it into the shopping basket.

[0086] When the user (customer) scans the code symbol attached to the product (step S1), the registration processing unit 501 performs a registration process (step S2). That is, the registration processing unit 501 decodes the code symbol and transmits the obtained product code to the application server 1.

[0087] When the registration processing unit 101 of the application server 1 receives the product code from the mobile POS 5, it acquires the product information corresponding to the product code from the product master 192 (step S3). Subsequently, the registration processing unit 101 registers the product information acquired in step S3 as transaction information 195 (step S4), that is, stores it in the storage unit 19. Also, the registration processing unit 101 transmits the registered transaction information 195 to the mobile POS 5 (step S5).

[0088] Steps S1 to S5 are repeatedly performed during the shopping of the user (customer) of the mobile POS 5.

[0089] Next, when the user (customer) operates the operation input unit 56 to request the start of a chat (step S11), the dialogue processing unit 502 requests the application server 1 to start AI cooperation (step S12). In the application server 1 that has received the request, the AI cooperation unit 102 requests the artificial intelligence server 3 to start cooperation (step S13). In the artificial intelligence server 3 that has received the request, the cooperation unit 302 prepares for cooperation with the mobile POS 5 and the application server 1.

[0090] After the chat starts, the user (customer) asks a question to the mobile POS 5 (step S21). The mobile POS 5 receives the input of the question in voice or text through a microphone or a touch panel as the operation input unit 56.

[0091] The dialogue processing unit 502 transmits the information indicating the user's question captured via the operation input unit 56 to the application server 1 (step S22).

[0092] When the control unit 10 of the application server 1 receives the information indicating the question from the mobile POS 5, as the AI cooperation unit 102, it transmits the information indicating the question to the artificial intelligence server 3 (step S23).

[0093] When the control unit 30 of the artificial intelligence server 3 receives the information indicating the user's question from the application server 1, as the cooperation unit 302, it requests the information necessary to answer the question from the application server 1 (step S31).

[0094] When the control unit 10 of the application server 1 receives the information request from the artificial intelligence server 3, as the AI cooperation unit 102, it acquires the information that matches the request (step S32). This information acquisition is mainly extracted from the product master 192, purchase history 193, and transaction information 195 stored in its own storage unit 19.

[0095] The AI cooperation unit 102 transmits the information acquired in step S32 to the artificial intelligence server 3 as a response to the request (step S33).

[0096] When the control unit 30 of the artificial intelligence server 3 receives the information in step S33, as the cooperation unit 302, it generates a proposal (step S41). The learning result data 400 is used for generating the proposal, and a proposal based on the learning result data 400 is generated. The cooperation unit 302 transmits the information indicating the generated proposal to the application server 1 (step S42).

[0097] Here, if the appearance probabilities of each word exemplified in FIG. 9 are always used as they are, there is a risk that biases such as always adopting materials or additional conditions with high appearance probabilities will occur in the response sentence. In this case, proposals that stimulate new products or the customer's creative desire will not be generated, which is not preferable. Therefore, for example, by using (1 - appearance probability) as the appearance probability at a predetermined frequency, etc., to change the output of the response sentence, it becomes easier to generate proposals that stimulate new products or the customer's creative desire, which is more suitable.

[0098] When the control unit 10 of the application server 1 receives the information of step S42 from the artificial intelligence server 3, as the AI cooperation unit 102, it transfers (transmits) the information to the mobile POS 5 (step S43). The dialogue processing unit 502 of the mobile POS 5 that has received this displays the information on the display unit 55 (step S44).

[0099] Steps S21 to S44 are repeated during the chat between the mobile POS 5 and the user (customer).

[0100] Here, FIG. 13 is a diagram schematically showing an example of a chat. This diagram shows the interaction (alternating statements) between the customer and the system of this embodiment. Note that the statements with odd numbers are the customer's questions, and the statements with even numbers are the answers by the system. Also, the customer's questions are due to step S21. The system's answers are due to step S44. The questions include text input or voice input, and the answers include text output or voice output.

[0101] When the customer asks the first question "What can be made in the shopping basket?" (steps S21 to S23), in response, the system generates a proposal (response sentence) through the processing of steps S31 to S41. The information transmitted in step S33 includes transaction information 195 (FIG. 4) indicating the contents of the customer's shopping basket, purchase history 193 (FIG. 5), and product information extracted from the product master 192 (FIG. 3). In the illustrated example, the system recognizes from the information in step S33 that the materials "ka", "sa", and "ta" are in the customer's shopping basket.

[0102] Next, based on the result of, for example, matching transaction information 195 with learning result data 400, the system returns the second answer "XX, YY, and chairs can be made." (Steps S42 to S44). In the illustrated example, in step S41, the system obtains the category "chair" of the products "a1" to "a4" that include any of the products in the shopping basket as materials by, for example, matching transaction information 195 (Fig. 4) with product information 401 (Fig. 7) and tool information 402 (Fig. 8).

[0103] On the other hand, when the customer asks the third question "Tell me about the chair." (Steps S21 to S23), the system performs the processing of steps S31 to S41 again as necessary to generate the next proposal (answer). For example, based on transaction information 195 and learning result data 400, the system returns the fourth answer "The chair that can be made with the current materials is 'a1'. Tools 'A' and 'B' are required." (Steps S42 to S44). At this time, it is preferable for the system to show the expected completion state of the assumed materials in an image.

[0104] Next, when the customer asks the fifth question "I have the tools. A more luxurious chair would be nice." (Steps S21 to S23), the system performs the processing of steps S31 to S41 again as necessary to generate the next proposal.

[0105] For example, the system obtains information on the product "a2" that has the same category "chair" as the proposed product "a1", has the additional condition of "luxury", and includes the material "shi" that exists in the customer's purchase history 193 by referring to product information 401, additional information 403, and purchase history 193. Then, the system returns the sixth answer "If you purchase materials'shi' and 'chi', a luxurious 'a2' can be made. Since you have purchased'shi' in the past transactions, it is recommended." (Steps S42 to S44). At this time, it is preferable for the system to show the expected completion state of the assumed materials in an image.

[0106] In response, if the customer asks the 7th question, "I didn't like 'Shi' very much when I purchased it before. I prefer a high-class chair made of other materials." (Steps S21 to S23), the system will perform the processing of Steps S31 to S41 again as necessary to generate the next proposal (answer).

[0107] For example, the system refers to the combination of materials of the product being a chair and the additional information 403, and obtains the information of 'Se' whose appearance probability of the additional condition 'high-class feeling' is next highest after 'Shi'. Then the system returns the 8th answer, "By adding the material 'Se', 'A4' can be made. It has a high-class feeling and the working time is standard." (Steps S42 to S44). At this time, it is preferable for the system to show the expected completion state of the assumed materials in an image.

[0108] In response, if the customer asks the 9th question, "I decide on 'A4'. Please teach me the details of the location of the materials and how to make them." (Steps S21 to S23), the system will perform the processing of Steps S31 to S41 again as necessary to generate the next proposal (answer).

[0109] For example, in the 10th answer based on the learning result data 400, the system presents the information indicating the display location of the product used as the material and the information indicating how to make the product by displaying it on the display unit 55, etc. (Steps S42 to S44). The user saves the presented information as appropriate if necessary.

[0110] More specifically, in the above proposal (answer), the system shows the materials of 'A4', etc., and obtains the display location of items not included in the transaction information 195 and the purchase history 193 from the product master 192 (Figure 3), and notifies the customer in Steps S42 to S44.

[0111] After that, the customer goes to the presented display location, for example, picks up the desired product, and scans the code symbol (Step S1). In response, the system registers the product information in the transaction information 195 (Steps S2 to S5).

[0112] In addition, if the store provides a method that allows customers to make purchases without going to the display location to pick up the products, the operation input unit 56 of the mobile POS 5 that presents the information in steps S42 to S44 may accept the product registration operation. Examples of the above method include a service where the store clerk delivers the corresponding product to the position indicated by the information of the mobile POS 5 after product registration or at the time of settlement, or a home delivery service.

[0113] Return to FIG. 12. When the user (customer) operates the operation input unit 56 to request the end of the chat (step S51), the dialogue processing unit 502 requests the application server 1 to end the AI cooperation (step S52). In the application server 1 that has received the request, the AI cooperation unit 102 requests the artificial intelligence server 3 to end the cooperation (step S53). In the artificial intelligence server 3 that has received the request, the cooperation unit 302 ends the cooperation with the mobile POS 5 and the application server 1.

[0114] Next, when the user (customer) operates the operation input unit 56 to request the settlement of the products placed in the shopping basket (that is, the products registered in the transaction information 195) (step S61), the settlement processing unit 503 requests the application server 1 to execute the settlement processing (step S62). In the application server 1 that has received the request, the settlement processing unit 103 performs the settlement processing (payment of the price by the customer, processing related to settlement) and updates the transaction information 195 (step S63).

[0115] The settlement processing unit 103 transmits the settled transaction information 195 to the mobile POS 5 (step S64). The settlement processing unit 503 of the mobile POS 5 that has received this displays information indicating that the settlement has been completed on the display unit 55 (step S65).

[0116] In addition, after step S64, the settlement processing unit 103 transmits (feeds back) the settled transaction information 195 to the artificial intelligence server 3 (step S71). In the artificial intelligence server 3 that has received this, the update processing unit 303 updates the learning result data 400 (step S72).

[0117] As described above, according to this embodiment, by causing an artificial intelligence to learn existing information that is the source of the information to be obtained and using the artificial intelligence learned based on the existing information, it is possible to propose to the user a new and creative product that does not depend on existing kits or manufacturing methods, or a product that stimulates creativity.

[0118] More specifically, according to this embodiment, for example, it is possible to propose arranging existing information, such as changing the color of an existing product, adding decorations, or changing the dimensional balance described in the original manufacturing method. Further, according to this embodiment, it is possible to propose adjustments to ease of manufacture (difficulty) and cost perception, and changes to the impression of the appearance.

[0119] Further, according to this embodiment, for example, it is possible to create a different product from existing materials. Specifically, for example, it is possible to propose information for manufacturing a desk using a kit for manufacturing a chair and a kit for manufacturing a shelf. In this case, it is preferable that the control unit 30 of the artificial intelligence server 3 generates an image predicting the completed state and attaches it to the proposal.

[0120] Furthermore, according to this embodiment, it is also possible to propose a completely new product based on the user's purchase history 193, transaction information 195, product master 192, and the like. For example, among the products handled by the store, by combining known materials (such as lumber) whose information such as dimensions and materials is known, it is possible to generate a manufacturing method and a predicted completion drawing of a non-existing article, and provide it as information on a new product.

[0121] As described above, according to such an embodiment, it is possible to provide a proposal that stimulates creativity to the user (customer), and it is also possible to guide the materials and tools required for manufacturing the product without omission and promote purchases. Therefore, it is possible to prevent inconveniences such as the user's motivation being diminished due to purchasing additional missing materials and tools, and improve the user's satisfaction.

[0122] The above-described embodiments can be appropriately modified and implemented by changing a part of the configuration or functions of each of the above-described devices.

[0123] For example, in the above embodiment, various functional units (registration processing unit 501, dialogue processing unit 502, accounting processing unit 503, registration processing unit 101, AI cooperation unit 102, accounting processing unit 103, learning processing unit 301, cooperation unit 302, update processing unit 303) are provided by three devices (mobile POS 5, application server 1, artificial intelligence server 3). However, the implementation is not limited to this. For example, the mobile POS 5 may be equipped with all the functions realized by the above various functional units. Further, the application server 1 and the artificial intelligence server 3 may be configured as one server, and the above various functional units may be borne by the mobile POS 5 and the one server.

[0124] Note that the programs executed by each device in the above-described embodiments are provided by being pre-installed in a ROM or the like. The programs executed by each device in the above-described embodiments may be configured to be recorded on a computer-readable non-transitory recording medium such as a CD-ROM, a flexible disk (FD), a CD-R, a DVD (Digital Versatile Disk) in an installable format or an executable format file and provided.

[0125] Furthermore, the programs executed by each device in the above-described embodiments may be configured to be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Also, the programs executed by each device in the above-described embodiments may be configured to be provided or distributed via a network such as the Internet.

[0126] Although some embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, changes, and combinations can be made without departing from the gist of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalent scope thereof.

Explanation of Signs

[0127] 1 … Application server, 10… Control unit, 101… Registration processing unit, 102… AI cooperation unit, 103… Accounting processing unit, 11… CPU, 12… ROM, 13… RAM, 14… Communication unit, 19… Storage unit (an example of the first storage means and the second storage means), 191… Program, 192… Product master, 193… Purchase history, 195… Transaction information, 2 … Network, 200… Teacher data, 201… Off-the-shelf information, 202… Knowledge information, 3 … Artificial intelligence server, 30… Control unit (an example of the generation means), 301… Learning processing unit, 302… Cooperation unit, 303… Update processing unit, 31… CPU, 32… ROM, 33… RAM, 34… Communication unit, 39… Storage unit, 391… Program, 400… Learning result data, 401… Product information, 402… Tool information, 403… Additional information, 5 … Mobile POS, 50… Control unit, 501… Registration processing unit, 502… Dialogue processing unit, 503… Accounting processing unit, 51… CPU, 52… ROM, 53… RAM, 54… Communication unit, 55… Display unit (an example of the presentation means), 56… Operation input unit (an example of the reception means), 57... Audio output unit, 58... Imaging unit, 59... Memory unit, 591... Program.

Prior Art Documents

Patent Documents

[0128]

Patent Document 1

Claims

1. A first storage means for storing product information which is information on products handled by a store; A second storage means for storing transaction information which is the product information registered as a transaction target; A reception means for receiving an input of a customer's request; A generation means for generating a new product that can be produced using at least a part of the transaction information based on learning result data generated by machine learning using information on an existing product as teacher data; A presentation means for presenting the information output by the generation means in a recognizable manner to the customer; A shopping support system comprising the above.

2. In generating a new product based on the learning result data, the generation means generates a new product that can be produced using, in addition to the transaction information, purchase history information which is information on products purchased by the customer who conducts the transaction in the past and at least a part of the product information stored in the first storage means. The shopping support system according to Claim 1. The shopping support system according to Claim 1.

3. The reception means and the presentation means exchange information in a dialogue format using voice or characters. The generation means generates information on materials, tools, and manufacturing methods for manufacturing the new product, and the presentation means presents the materials, tools, and manufacturing methods in the dialogue format. The shopping support system according to Claim 1.

4. The shopping support system according to Claim 1, further comprising a learning processing unit that obtains the learning result data by machine learning at least one of information indicating how to make an existing product, information indicating tools used for manufacturing the product, and knowledge information of store clerks in the store as the teacher data.

5. A reception means for receiving an input of a customer's request; A generation means for generating a new product that can be produced using at least a part of the transaction information which is the product information registered as a transaction target among the product information which is information on products handled by a store based on learning result data generated by machine learning using information on an existing product as teacher data; A presentation means for presenting the information output by the generation means in a recognizable manner to the customer; A sales data processing apparatus comprising the above.

6. A computer included in a shopping support system including a first storage means for storing product information which is information on products handled by a store and a second storage means for storing transaction information which is the product information registered as a transaction target, A reception means for receiving an input of a customer's request; A generation means for generating a new product that can be produced using at least a part of the transaction information based on learning result data generated by machine learning with information on existing products as teacher data; A presentation means for presenting the information output by the generation means in a recognizable manner to the customer; A shopping support program for functioning as.

Images