system

Abstract

Provide a system. means for inputting an alarm message; means for receiving the alarm message, and tokenizing and extracting keywords from the alarm message, the means comprising a natural language processing engine; means including a machine learning model for classifying alarm types based on the keywords; means for searching a database for an appropriate measure for the type of alarm and displaying the result on a user interface; A system including:

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 an instruction sentence related to a description of 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] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Alarms frequently occur during base station construction, but the content is provided entirely in English, making it difficult to understand. While there is a need to properly analyze the alarm content, identify the cause of the problem, and quickly implement countermeasures, there is a lack of systems to efficiently do this. Therefore, there is a need to provide a system that automates the analysis of alarm messages and proposes countermeasures, allowing users to quickly take appropriate action. [Means for solving the problem]

[0005] The present invention provides a system that inputs alarm messages through a user interface, tokenizes them using a natural language processing engine, extracts keywords, and classifies the alarm type based on the results using a machine learning model. Furthermore, the system searches a database for appropriate countermeasures for the classified alarm and displays the results on the user interface. The system also includes a log storage means for recording received alarm message data and a function for analyzing the cause of the alarm based on the keywords extracted by the natural language processing engine. This allows users to quickly understand the content and cause of the alarm and take appropriate countermeasures.

[0006] A "user interface" is a software or hardware element that contains input and output means by which a user interacts with a system.

[0007] "Alarm message" refers to a notification message regarding abnormalities or failures that occur during the construction or operation of a base station.

[0008] "Reception" refers to the action of a terminal or server receiving data sent from an external source.

[0009] "Tokenization" refers to the process of dividing text into small units (tokens) in natural language processing.

[0010] "Keywords" refer to important words or phrases extracted during analysis of alarm messages.

[0011] A "natural language processing engine" is a general term for software that analyzes text data and performs tasks such as tokenization and keyword extraction.

[0012] A "machine learning model" refers to an algorithm that learns from data and makes classifications and predictions.

[0013] "Classification" refers to the process of grouping received data according to certain criteria.

[0014] A "database" is a system that stores information in an organized manner and makes it easy to search and retrieve.

[0015] "Logging" refers to recording received data and storing it for future reference.

[0016] "Countermeasure" refers to the corrective action or course of action taken in response to a particular alarm or anomaly.

[0017] "Analysis" refers to the process of examining specific data or information in detail to understand its content and structure. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 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. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 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. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 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. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0019] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0020] First, the terms used in the following description will be explained.

[0021] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single 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), and an APU (Accelerated Processing Unit).

[0022] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0023] 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.

[0024] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0025] 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. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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 WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] 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.

[0031] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. 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.

[0032] 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 (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. 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 process 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.

[0036] 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.

[0037] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. 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.

[0038] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. 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] This invention relates to a system that inputs alarm messages that occur during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures. This system consists of a Uzentoface, a natural language processing engine on the server, a machine learning model, and a database.

[0040] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5." The terminal then sends this message to the server.

[0041] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[0042] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[0043] The detected countermeasures are displayed to the user through the user interface. For example, a countermeasure such as "Check the backup power supply and replace it if necessary" is displayed on the screen. As a result, the user can quickly understand the type and cause of the alarm and take appropriate countermeasures.

[0044] The specific program processing flow is explained below. When a user inputs an alarm message and sends it to the server via the terminal, the server receives the message and saves it as a log. The message is then analyzed by a natural language processing engine, and tokenization and keyword extraction are performed. The extracted keywords are classified using a machine learning model, and appropriate countermeasures are searched for in the database based on the obtained alarm type. The searched countermeasures are finally displayed to the user via the terminal.

[0045] This system solves the problem of alarm messages being difficult to understand, allowing users to respond appropriately and quickly. For example, if the message "Power supply failure detected at base station 5" is input, the system will classify this as a "power supply failure" and suggest the countermeasure of "checking the backup power supply and replacing it if necessary." As a result, users can carry out their work more efficiently.

[0046] The processing flow will be explained below.

[0047] Step 1:

[0048] The user inputs an alarm message into the terminal, for example, "Power supply failure detected at base station 5."

[0049] Step 2:

[0050] The terminal sends the entered alarm message to the server. The message is sent to the server as a character string.

[0051] Step 3:

[0052] The server receives alarm messages sent from the terminals and records the received messages in a database as a log.

[0053] Step 4:

[0054] The server passes the received message to a natural language processing engine (NLP engine), which tokenizes the message and extracts keywords and important phrases.

[0055] For example, keywords such as "Power supply failure" and "base station 5" are extracted.

[0056] Step 5:

[0057] The server inputs the extracted keywords into a machine learning model, which then classifies the alarm type based on these keywords.

[0058] For example, it may be classified as "power supply failure" based on keywords.

[0059] Step 6:

[0060] Based on the alarm classification, the server searches for the appropriate countermeasure from a database that records past response cases and corrective actions.

[0061] Step 7:

[0062] The server converts the detected countermeasure information into a format for display on the user interface. For example, a countermeasure such as "Check the standby power supply and replace it if necessary" is selected.

[0063] Step 8:

[0064] The server generates and sends formatted countermeasure information to the terminal.

[0065] Step 9:

[0066] The terminal displays the countermeasure information received from the server to the user. The user can see information such as "Alarm type: Power supply failure" and "Corrective action: Check the backup power supply and replace it if necessary" on the screen.

[0067] Step 10:

[0068] The user can take appropriate action based on the displayed instructions, thereby quickly implementing measures to resolve the cause of the alarm.

[0069] Example 1

[0070] Next, a description will be given of 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."

[0071] There is a need for a system that can quickly and accurately analyze the contents of alarm messages that occur during base station construction and propose appropriate countermeasures. However, conventional systems have slow processes for understanding alarm messages and proposing countermeasures, making efficient problem solving difficult. For this reason, there is a need to consistently automate the rapid classification of alarms and the proposal of countermeasures, thereby reducing the burden on users.

[0072] 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.

[0073] In this invention, the server includes means for a user to input an alarm message via a terminal, means for receiving the alarm message and saving it as a log, and means for tokenizing the alarm message using a natural language processing engine and extracting keywords, thereby enabling the alarm message to be quickly analyzed and appropriate countermeasures to be presented.

[0074] A "user" is a person or organization that utilizes the system to enter alarm messages.

[0075] A "terminal" is an electronic device that a user uses to input an alarm message and transmit the message to a server.

[0076] An "alarm message" is a message that indicates an abnormality or problem that occurs during the construction of a base station.

[0077] A "server" is a central computer system that receives, analyzes, and processes alarm messages.

[0078] A "log" is a recording file or database that stores received alarm messages.

[0079] A "natural language processing engine" is a software program that analyzes alarm messages and performs tokenization and keyword extraction.

[0080] "Tokenization" is a process of dividing an alarm message into units such as words and phrases.

[0081] "Keywords" are important words or phrases extracted from alarm messages.

[0082] A "machine learning model" is an algorithm for classifying alarm types based on extracted keywords.

[0083] "Alarm type" refers to a category of abnormalities or problems classified based on keywords.

[0084] A "database" is an information collection system that stores past data and knowledge bases and searches for appropriate countermeasures.

[0085] "User interface" refers to the screens and operating means through which a user interacts with a system.

[0086] "Countermeasures" are specific actions or treatments recommended for a classified alarm.

[0087] This invention relates to a system that inputs alarm messages that occur during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures. This system consists of a user interface, a natural language processing engine on the server, a machine learning model, and a database.

[0088] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5.", and the terminal then sends this message to the server.

[0089] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[0090] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[0091] The detected countermeasures are displayed to the user through the user interface. For example, a countermeasure such as "Check the backup power supply and replace it if necessary" is displayed on the screen. As a result, the user can quickly understand the type and cause of the alarm and take appropriate countermeasures.

[0092] For example, if a user inputs the message "Power supply failure detected at base station 5," the system will classify this as a "power supply failure" and suggest a solution: "Check the backup power supply and replace it if necessary." As a result, the user can solve the problem appropriately and quickly.

[0093] Hardware and software used

[0094] The terminals are general-purpose personal computers or smartphones. Secure HTTPS is used as the communication protocol. A high-performance computer server is recommended. Natural language processing engines such as spaCy and NLTK can be used. Machine learning models are built using scikit-learn and TENSORFLOW (registered trademark). The database uses widely used database management systems such as MySQL (registered trademark) and PostgreSQL.

[0095] Prompt Sentence Examples

[0096] User Prompt: "Power supply failure detected at base station 5."

[0097] Expected AI Response: "Category: Power Supply Failure. Recommended Action: Check backup power supply and replace if necessary."

[0098] This system allows users to quickly understand the content of alarm messages and take appropriate action. Specifically, if the prompt "Power supply failure detected at base station 5" is entered, the system will classify this as a "power supply failure" and provide the user with appropriate countermeasures. This improves the work efficiency of system users.

[0099] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0100] Step 1:

[0101] The user inputs an alarm message through the terminal. Specifically, the user inputs "Power supply failure detected at base station 5." into the text box on the terminal. When the user sends the input, the terminal sends this message to the server. The input is the alarm message in text format, and the output is the data to be sent to the server.

[0102] Step 2:

[0103] The terminal sends the alarm message to the server. Specifically, the terminal sends the entered message to the server using the secure HTTPS protocol. The input is the text message entered by the user, and the output is the raw alarm message that arrives at the server.

[0104] Step 3:

[0105] The server saves alarm messages in a log. Specifically, the server saves the messages it receives in a database. At this time, the message timestamp and sender information are also saved. The input is the alarm message sent from the terminal, and the output is the log entry saved in the database.

[0106] Step 4:

[0107] The server parses the message using a natural language processing engine. Specifically, the server passes the stored message to a natural language processing engine (e.g., spaCy or NLTK) for tokenization and parsing. The input is the stored alarm message, and the output is a tokenized keyword list.

[0108] Step 5:

[0109] A natural language processing engine tokenizes the message and extracts keywords, such as "Power supply failure" and "base station 5." The input is the text message waiting to be parsed, and the output is the extracted keywords.

[0110] Step 6:

[0111] The server inputs the extracted keywords into a machine learning model. Specifically, the server inputs the keyword list into a machine learning model (e.g., a model built with scikit-learn or TensorFlow) to classify the alarm type. The input is the extracted keywords, and the output is the classified alarm type (e.g., "power supply failure").

[0112] Step 7:

[0113] The machine learning model classifies alarm types, specifically by categorizing keywords into categories such as "power supply failure." The input is a list of keywords, and the output is the alarm type category.

[0114] Step 8:

[0115] The server searches the database for appropriate countermeasures. Specifically, based on the classified alarm type, the server refers to past data and a knowledge base in the database to search for appropriate countermeasures. The input is the classified alarm type, and the output is a list of countermeasures (for example, "Check the backup power supply and replace it if necessary").

[0116] Step 9:

[0117] The server displays the search results to the user through a user interface. Specifically, the countermeasures searched by the server are displayed on the user's device through the user interface. The input is a list of countermeasures, and the output is the displayed countermeasure information. The user can check this and quickly implement countermeasures.

[0118] (Application example 1)

[0119] Next, a description will be given of Application 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."

[0120] When an abnormality occurs in equipment at a manufacturing site, it is necessary to quickly identify the cause of the abnormality and take appropriate measures. However, it is difficult for on-site workers to respond immediately to all abnormalities, which results in a decrease in production efficiency. To solve this problem, a system that automatically detects abnormalities and proposes countermeasures is needed.

[0121] 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.

[0122] In this invention, the server includes means having a user interface for inputting alarm messages, means having a natural language processing engine that receives the alarm messages, tokenizes them, and extracts keywords, means including a machine learning model that classifies alarm types based on the keywords, means for searching a database for appropriate countermeasures for the alarm types and displaying the results on the user interface, means for transmitting alarm messages from devices in the factory that are equipped with anomaly detection sensors, and means for analyzing the alarm messages and suggesting appropriate countermeasures to factory workers. This makes it possible to quickly identify the cause of an anomaly and take appropriate countermeasures.

[0123] An "alarm message" is a message that notifies of an abnormality detected by an abnormality detection sensor in a factory.

[0124] A "user interface" is an interface that allows factory workers to access the system and perform input and output.

[0125] The "natural language processing engine" is an engine that analyzes input alarm messages, tokenizes them, and extracts keywords.

[0126] "Tokenization" is the process of dividing and extracting words and phrases from a text.

[0127] "Keywords" are important words or phrases extracted from the input alarm message.

[0128] A "machine learning model" is an algorithm that learns from data and classifies alarm types.

[0129] A "database" is a system for storing and searching past data and countermeasures.

[0130] An "abnormality detection sensor" is a sensor installed to detect abnormalities in factory equipment.

[0131] "Factory workers" are workers who operate production equipment in factories and perform maintenance and inspections.

[0132] "Countermeasure proposal" is the act of proposing an appropriate solution to a detected abnormality.

[0133] A "server" is a computer system that receives alarm messages and performs analytical processing.

[0134] The present invention relates to a system that detects and analyzes abnormalities that occur in a factory and proposes appropriate countermeasures. This system is composed of a user interface, a natural language processing engine installed on a server, a machine learning model, and a database. A specific embodiment of this system is described below.

[0135] First, when an anomaly detection sensor installed in the factory detects an abnormality, an alarm message is generated. This message might say, for example, "The conveyor belt has stopped." This alarm message is sent from a terminal in the factory to the server.

[0136] The server stores the received alarm messages as logs. These logs contain the data needed for later analysis. A natural language processing engine then analyzes the received messages, tokenizing them and extracting keywords. In this example, keywords such as "conveyor belt" and "stop" are extracted.

[0137] The extracted keywords are input into a machine learning model, which classifies the alarm type. For example, in this case, it is classified as "machine stopped." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base. The countermeasures found are displayed to factory workers through a user interface. For example, a countermeasure such as "Check the motor and replace it if necessary" is displayed on the screen.

[0138] This system enables factory workers to quickly understand the nature and cause of abnormalities and take appropriate measures. Specifically, factory workers can access the system using tablet devices and respond by following the instructions provided by the server.

[0139] The hardware used includes anomaly detection sensors, factory terminals (e.g., tablets), and servers, while the software used includes natural language processing engines (e.g., spaCy and nltk), machine learning models (e.g., scikit-learn and TensorFlow), and databases (e.g., SQLite and MySQL).

[0140] For example, if an alarm message such as "The conveyor belt has stopped" is input, spaCy extracts the keywords "conveyor belt" and "stopped," and scikit-learn classifies it as "machine stopped," and suggests appropriate countermeasures from the database, such as "check the motor and replace it if necessary." In this way, factory workers can immediately take countermeasures.

[0141] An example of a prompt sentence is "The conveyor belt has stopped." By inputting this prompt sentence into the system, the system is designed to suggest the optimal solution.

[0142] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0143] Step 1:

[0144] The abnormality detection sensor detects an abnormality.

[0145] Specific operation: Anomaly detection sensors in the factory detect machine stoppages or abnormal operation.

[0146] Input: Signal from anomaly detection sensor.

[0147] Output: An abnormal message is generated (e.g., "The conveyor belt has stopped.").

[0148] Step 2:

[0149] Send an abnormality message to the server.

[0150] Specific operation: Upon receiving a signal from an anomaly detection sensor, a terminal in the factory generates an alarm message and sends it to a server via the Internet.

[0151] Input: Anomaly message (e.g. "The conveyor belt has stopped.").

[0152] Output: The server receives the abnormal message.

[0153] Step 3:

[0154] The server stores the error message in a log.

[0155] Specific operation: The server records the received abnormal messages in a text file or database.

[0156] Input: Anomaly message (e.g. "The conveyor belt has stopped.").

[0157] Output: Abnormal messages are saved to a log file or database.

[0158] Step 4:

[0159] The server analyzes the abnormal message using a natural language processing engine.

[0160] Specific operation: Tokenize the anomaly message using a natural language processing engine (e.g., spaCy) and extract keywords.

[0161] Input: Anomaly message from a log file or database (e.g., "The conveyor belt has stopped.").

[0162] Output: Extracted keywords (e.g. "conveyor belt", "stop").

[0163] Step 5:

[0164] The server uses a machine learning model to classify the alarm type.

[0165] Specific operation: Using a machine learning model (e.g., scikit-learn), classify the alarm type based on the extracted keywords.

[0166] Input: Extracted keywords (e.g., "conveyor belt", "stop").

[0167] Output: Type of alarm (e.g. "Machine Stop").

[0168] Step 6:

[0169] The server searches the database for an appropriate solution.

[0170] Specific operation: Based on the classified alarm type, the system searches the database for past data and knowledge base to find the appropriate countermeasure.

[0171] Input: Type of alarm (e.g. "Machine Stop").

[0172] Output: Appropriate measures (e.g. "Check motor and replace if necessary").

[0173] Step 7:

[0174] The server displays the appropriate measures on the user interface.

[0175] Specific operation: After finding an appropriate countermeasure, it displays it on the user interface used by factory workers. If necessary, it outputs instructions and procedures.

[0176] Input: Appropriate action (e.g. "Check motor and replace if necessary").

[0177] Output: The countermeasures are displayed in the user interface.

[0178] Furthermore, an emotion engine that estimates the user's emotion may be 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.

[0179] This invention relates to a system that inputs alarm messages generated during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures, and also combines them with an emotion engine that recognizes user emotions. This system is composed of a user interface, a natural language processing engine on a server, a machine learning model, an emotion engine, and a database.

[0180] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5." The terminal then sends this message to the server.

[0181] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[0182] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[0183] Additionally, an emotion engine is also in operation to recognize the emotions the user is feeling when they type. The emotion engine analyzes the emotions from the text the user enters and can identify emotions such as "anxiety" or "tension." Based on this information, the server adjusts the way it presents solutions. For example, if the user is feeling very stressed, it will use calmer, more positive expressions.

[0184] The system then displays advice tailored to the user based on the searched solutions and the emotion engine results. For example, if the solution "Check the backup power supply and replace it if necessary" is found, the system also displays encouraging words such as "Remain calm and continue working. The problem can be solved."

[0185] The specific program processing flow is explained below. When a user inputs an alarm message and sends it to the server via their terminal, the server receives the message and saves it as a log. The message is then analyzed by a natural language processing engine, and tokenization and keyword extraction are performed. The extracted keywords are classified using a machine learning model, and appropriate countermeasures are searched for in a database based on the obtained alarm type. At the same time, the emotion engine analyzes the user's emotions. The way the countermeasures are presented is adjusted based on the emotional information. Finally, countermeasure information and an encouraging message corresponding to the user's emotions are displayed to the user via their terminal.

[0186] This system solves the problem of difficulty in understanding alarm messages and enables appropriate and prompt responses according to the user's emotional state. For example, if the message "Power supply failure detected at base station 5" is input, the system will classify this as a "power supply failure" and provide measures such as "check the backup power supply and replace it if necessary," as well as an encouraging message such as "please stay calm and continue working. The problem can be solved." As a result, the user can carry out their work efficiently and with less mental strain.

[0187] The processing flow will be explained below.

[0188] Step 1:

[0189] The user inputs an alarm message into the terminal, for example, "Power supply failure detected at base station 5."

[0190] Step 2:

[0191] The terminal sends the entered alarm message to the server. The message is sent to the server as a character string.

[0192] Step 3:

[0193] The server receives alarm messages sent from the terminals and records the received messages in a database as a log.

[0194] Step 4:

[0195] The server passes the received message to a natural language processing engine (NLP engine), which tokenizes the message and extracts keywords and important phrases, such as "Power supply failure" and "base station 5."

[0196] Step 5:

[0197] The server inputs the extracted keywords into a machine learning model, which then classifies the alarm type based on these keywords. For example, the alarm may be classified as "power supply failure" based on the keywords.

[0198] Step 6:

[0199] Based on the alarm classification, the server searches for the appropriate countermeasure from a database that records past response cases and corrective actions.

[0200] Step 7:

[0201] To recognize the emotions expressed when a user inputs an alarm message, the server runs an emotion engine. The emotion engine extracts emotion data from the input message and user operations, and analyzes the user's emotional state, such as "anxiety" or "tension."

[0202] Step 8:

[0203] The server analyzes the emotional information recognized by the emotion engine and adjusts the way the countermeasures are presented. If the user is feeling strong stress, it will use calmer and more positive expressions.

[0204] Step 9:

[0205] The server converts the detected countermeasure information and adjusted information based on it into a format for display in the user interface. For example, in addition to the countermeasure "Check the backup power supply and replace it if necessary," it generates an encouraging message such as "Please stay calm and continue working. The problem can be solved."

[0206] Step 10:

[0207] The server sends the generated formatted countermeasure information and the adjusted information to the terminal.

[0208] Step 11:

[0209] The device receives countermeasure information from the server and displays advice based on the user's emotion. The user can see information on the screen such as "Alarm type: Power supply failure," "Corrective action: Check the backup power supply and replace it if necessary," and "Please remain calm and continue working. The problem can be solved."

[0210] Step 12:

[0211] The user can take appropriate action based on the displayed instructions and encouraging messages, thereby quickly implementing measures to resolve the cause of the alarm.

[0212] Example 2

[0213] Next, a description will be given of 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."

[0214] Conventional alarm message analysis systems analyzed alarm messages and proposed appropriate countermeasures, but because they ignored the emotions felt by users when inputting messages, the countermeasures were sometimes not accepted or implemented smoothly. Furthermore, the lack of appropriate support for emotions such as anxiety and tension felt by users led to problems such as a heavy mental burden on users and reduced work efficiency. To solve these problems, a system that not only analyzes alarm messages but also takes into account the user's emotions is needed.

[0215] 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.

[0216] In this invention, the server includes means for saving alarm messages as logs, means having an emotion analysis engine for analyzing emotions from text entered by a user, and means for adjusting the manner in which countermeasures are presented based on the analyzed emotion information. This not only analyzes the contents of alarm messages and proposes appropriate countermeasures, but also adjusts the manner in which the countermeasures are presented according to the emotions felt by the user, thereby reducing mental burden and improving work efficiency.

[0217] An "alarm message" is a message used to report problems or abnormalities that occur on the system.

[0218] "User interface" refers to the screens and input devices that allow a user to interact with a system.

[0219] A "server" is a computer system that processes and stores data over a network.

[0220] A "natural language processing engine" is a technology that analyzes human language and allows computers to understand and analyze it.

[0221] "Tokenization" is the process of dividing a sentence into units such as words and phrases.

[0222] "Keywords" are words or phrases that are particularly important in a piece of text.

[0223] A "machine learning model" is an algorithm that learns from data and makes predictions and classifications.

[0224] A "database" is a system for efficiently managing and searching large amounts of data.

[0225] An "emotion analysis engine" is a technology that analyzes and identifies emotions from text data.

[0226] A "log" is recorded data that a system generates during operation.

[0227] The present invention is a system for analyzing alarm messages related to base station construction and proposing countermeasures, and is configured to propose countermeasures taking into consideration the user's emotions. This system is composed of a user interface, a natural language processing engine installed on a server, a machine learning model, a sentiment analysis engine, and a database.

[0228] First, the user inputs an alarm message through the terminal. For example, if the user inputs the message "Power supply failure detected at base station 5," the terminal sends this message to the server. The server receives the message and stores it in a database as a log. The stored message is then analyzed by a natural language processing engine and tokenized into words and phrases. In this case, keywords such as "Power supply failure" and "base station 5" are extracted.

[0229] The extracted keywords are then fed into a machine learning model, which uses a pre-trained algorithm to classify the alarm type based on the keywords. For example, it may be classified as "power supply failure." The server then searches a database for an appropriate countermeasure for the classified alarm. For example, it may find a countermeasure such as "check the backup power supply and replace it if necessary."

[0230] Furthermore, an emotion analysis engine analyzes the emotions from the text entered by the user. The emotion analysis engine identifies emotions such as "anxiety" or "tension" felt by the user. Based on this emotional information, the way the countermeasures are presented is adjusted. For example, if the user is feeling very anxious, the server will add encouraging words such as "Please stay calm and continue working. The problem can be solved."

[0231] Finally, the adjusted countermeasures and advice are displayed to the user via the terminal. As a specific example, if the user inputs the alarm message "Power supply failure detected at base station 5," the system will classify this as a "power supply failure" and display the countermeasure "Check the backup power supply and replace it if necessary," along with encouraging words such as "Please stay calm and work. The problem can be solved." An example of this prompt would be "Enter the alarm message and output the analysis results and countermeasures."

[0232] This system not only improves analysis accuracy compared to conventional systems, but also takes the user's emotions into consideration, reducing mental burden and enabling more appropriate countermeasures to be provided more quickly.

[0233] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0234] Step 1:

[0235] The user enters an alarm message

[0236] The user inputs an alarm message into the input field of the terminal. For example, the user inputs "Power supply failure detected at base station 5." This input is sent to the terminal as text data.

[0237] Step 2:

[0238] The device sends a message to the server

[0239] The device sends the input message to the server using an HTTP request, which is passed to the server as text data. At this time, the data is encoded into a common data format such as JSON.

[0240] Step 3:

[0241] The server receives the message and stores it in a log.

[0242] The server receives the HTTP request and stores it in temporary memory. Then, it saves the message as a log in a database. A database management system such as MySQL or PostgreSQL is used for saving. The input data is in text format, and the output is saved as a log file or a record in the database.

[0243] Step 4:

[0244] The server analyzes the message using a natural language processing engine

[0245] The server inputs the stored messages into a natural language processing engine. For example, it uses Python's NLTK library to tokenize the messages, which extracts keywords such as "Power supply failure" and "base station 5." The input is text data, and the output is a list of keywords.

[0246] Step 5:

[0247] The server classifies the alarm type using a machine learning model.

[0248] The server inputs the extracted keywords into a pre-trained machine learning model, for example, Scikit-learn, to classify the alarm type. The input is a list of keywords, and the output is an alarm type, such as "power supply failure."

[0249] Step 6:

[0250] The server searches the database for the appropriate solution

[0251] The server searches the database for appropriate countermeasures based on the classified alarm type. The database stores past alarm response history and a knowledge base. The input is the alarm type, and the output is a countermeasure such as "check the standby power supply and replace it if necessary."

[0252] Step 7:

[0253] The server analyzes the user's emotions using an emotion analysis engine.

[0254] The server uses a sentiment analysis engine to analyze the sentiment of the user's input text. For example, it uses Python's TextBlob library to identify emotions such as "anxiety" or "tension" that the user is feeling. The input is text data, and the output is an emotion label.

[0255] Step 8:

[0256] The server adjusts how the measures are expressed.

[0257] The server adjusts the presentation of the countermeasures it provides based on the analyzed emotion information. For example, if the user feels "anxious," it adds encouraging words such as "Please stay calm and work. The problem can be solved." The input is the emotion label and countermeasure data, and the output is the adjusted countermeasure message.

[0258] Step 9:

[0259] The device displays countermeasures and advice to the user

[0260] The terminal displays the countermeasure information sent from the server and the adjusted advice to the user. For example, the terminal displays "Check the backup power supply and replace it if necessary. Please stay calm and work. The problem can be solved." The input is the adjusted countermeasure message, and the output is the display on the user interface.

[0261] (Application example 2)

[0262] Next, a description will be given of Application 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."

[0263] In conventional factory robot operations, it has been difficult to immediately analyze error messages from the robot and provide appropriate countermeasures. Furthermore, if a robot operator feels anxious or stressed by an error message, there is a lack of means to provide support that responds to their emotions. This makes it difficult for operators to work efficiently and with peace of mind, and there is a risk of equipment stalling or incorrect operation.

[0264] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means having a user interface for inputting an alarm message; means equipped with a natural language processing engine that receives the alarm message, tokenizes it, and extracts keywords; means including a machine learning model that classifies the alarm type based on the keywords; means for searching a database for appropriate countermeasures for the alarm type and displaying the results on the user interface; means including an emotion analysis engine that recognizes the emotion of the user when entering the message; and means for adjusting the way advice is presented based on the results of the emotion analysis engine. This makes it possible to quickly and accurately analyze error messages from factory robots and provide appropriate countermeasures, as well as provide appropriate support according to the operator's emotional state.

[0265] Below are definitions of key terms found in the patent claims:

[0266] An "alarm message" refers to a notification sent when a device such as a robot or machine detects an abnormality or error.

[0267] "User interface" refers to the means or screen through which a user directly interacts with a computer system or application.

[0268] A "natural language processing engine" refers to the technology or system that enables a computer to understand and analyze human language and extract meaning.

[0269] "Tokenization" refers to the process of dividing a sentence into words and phrases.

[0270] A "keyword" is a word that has a particularly important meaning in a sentence or paragraph.

[0271] A "machine learning model" refers to an algorithm or system that learns patterns from data and uses those patterns to make inferences and classify new data.

[0272] A "database" refers to a system that systematically stores and manages data, allowing it to be quickly searched and used as needed.

[0273] An "emotion analysis engine" refers to a system that recognizes and analyzes emotions from user input, tone of voice, facial expressions, etc.

[0274] "Adjusting advice presentation" refers to changing the wording or presentation of advice or instructions provided depending on the user's emotional state.

[0275] This invention is a system that allows factory robot operators to analyze error messages from the robot and quickly provide appropriate countermeasures. Furthermore, it can improve work efficiency by recognizing the operator's emotions and providing appropriate support according to those emotions. This system includes the following components:

[0276] 1. Overall system configuration

[0277] The overall system configuration is as follows:

[0278] User interface: A means for operators to input error messages. This can be a factory robot control terminal, a smartphone, or a tablet device.

[0279] Server: A computer system that performs the main processing for analyzing error messages and providing appropriate countermeasures. The server is equipped with a natural language processing engine, machine learning models, a sentiment analysis engine, and a database.

[0280] Natural Language Processing Engine: A library for tokenizing error messages and extracting keywords, for example using Python's TextBlob library.

[0281] Machine learning model: A model for classifying error message types based on extracted keywords. This model is built using the sklearn library.

[0282] Sentiment analysis engine: An engine for recognizing emotions from operator input. For sentiment analysis, the transformers library is used, along with the BERT model.

[0283] Database: A source of information that accumulates solutions to error messages. This database stores past error messages and their corresponding solutions.

[0284] 2. System Operation

[0285] The flow for this system to work properly is as follows:

[0286] Alarm message entry and analysis

[0287] When an operator enters an error message on the factory robot's control terminal, the message is sent to the server. The server receives the message and saves it as a log. The natural language processing engine then analyzes the message and extracts the necessary keywords. For example, if the message entered is "Power supply failure detected at manufacturing robot 3," the words "Power supply failure" and "manufacturing robot 3" are extracted.

[0288] Error message classification

[0289] The extracted keywords are input into a machine learning model to classify the type of error message. For example, this one might be classified as "power supply failure." For each classified error message, the server accesses a database to search for the appropriate solution.

[0290] Sentiment analysis and advice provision

[0291] The emotion analysis engine analyzes the operator's emotions, identifying emotions such as "anxiety" or "stress" from input such as "This is really stressing me out." Based on this analysis result, the way countermeasures are presented is adjusted. For example, if the operator is feeling anxious, a positive message such as "Please stay calm and continue working. The problem can be solved" is added.

[0292] 3. Specific Examples

[0293] As a specific example, if an operator inputs the alarm message "Power supply failure detected at manufacturing robot 3" and the emotion input "This is really stressing me out," the system will classify this as a "power supply failure" and provide a solution: "Check the backup power supply and replace it if necessary." At the same time, it will also display an encouraging message: "Please stay calm and continue working. The problem can be solved."

[0294] Prompt Sentence Examples

[0295] "I'm experiencing a power supply failure. What's the best way to resolve this issue?"

[0296] "Please advise how to respond if users are feeling uneasy."

[0297] This system enables factory robot operators to deal with error messages quickly and appropriately, while at the same time reducing the psychological burden.

[0298] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0299] Step 1:

[0300] The user enters an error message.

[0301] Specific operation: The user uses the factory robot's management terminal or smartphone to enter an error message such as "Power supply failure detected at manufacturing robot 3."

[0302] Input: Error message

[0303] Output: The entered error message

[0304] Step 2:

[0305] The terminal sends an error message to the server.

[0306] Specific operation: The error message entered by the user is automatically sent from the management terminal to the server.

[0307] Input: Error message

[0308] Output: Error message sent to the server

[0309] Step 3:

[0310] The server receives the error message and stores it in a log.

[0311] Specific behavior: The server records the received error message in a database for later retrieval.

[0312] Input: The error message sent

[0313] Output: Error messages stored in the database

[0314] Step 4:

[0315] The server uses a natural language processing engine to tokenize the error message and extract keywords.

[0316] What it does: It uses Python's TextBlob library to analyze error messages and extract keywords such as "Power supply failure" and "manufacturing robot 3."

[0317] Input: Saved error message

[0318] Output: Extracted keywords

[0319] Step 5:

[0320] The server uses a machine learning model to classify the type of error message based on keywords.

[0321] Specific operation: The extracted keywords are input into a model built with the sklearn library and classified into error message types such as "power supply failure."

[0322] Input: Extracted keywords

[0323] Output: Classified error message types

[0324] Step 6:

[0325] The server searches the database for an appropriate solution based on the type of classified error message.

[0326] Specific operation: For an error message classified as "power supply failure," the database is searched for countermeasures such as "check the backup power supply and replace it if necessary."

[0327] Input: Classified error message type

[0328] Output: Measures found

[0329] Step 7:

[0330] The user inputs emotions.

[0331] Specific operation: The user inputs a feeling such as "This is really stressing me out" into the management terminal.

[0332] Input: Emotion input

[0333] Output: Input emotion text

[0334] Step 8:

[0335] The server analyzes the user's emotions using an emotion analysis engine.

[0336] Specific behavior: Analyzes input emotion text using the BERT model from the transformers library to identify emotions such as "anxiety" and "stress."

[0337] Input: Entered emotion text

[0338] Output: Parsed sentiment

[0339] Step 9:

[0340] The server adjusts the way advice is presented based on the emotions analyzed.

[0341] Specific action: If the analyzed emotion is "anxiety," add an encouraging message to the searched measures: "Please stay calm and work. The problem can be solved."

[0342] Input: Parsed sentiment, retrieved measures

[0343] Output: Adjusted advice

[0344] Step 10:

[0345] The server displays the tailored advice in a user interface.

[0346] Specific operation: Display the adjusted advice on the management terminal screen and provide the user with countermeasures and encouraging messages.

[0347] Input: Tailored Advice

[0348] Output: Advice displayed

[0349] 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 audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio 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 audio data.

[0350] 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 engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with 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.

[0351] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0352] [Second embodiment]

[0353] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0354] 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.

[0355] 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 WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0356] 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.

[0357] 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 audio in accordance with instructions from the processor 46.

[0358] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and 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 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0359] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0360] Fig. 4 shows an example of the 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.

[0361] 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.

[0362] 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.

[0363] In the smart glasses 214, the processor 46 performs the reception output process. 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.

[0364] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0365] This invention relates to a system that inputs alarm messages that occur during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures. This system consists of a Uzentoface, a natural language processing engine on the server, a machine learning model, and a database.

[0366] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5." The terminal then sends this message to the server.

[0367] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[0368] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[0369] The detected countermeasures are displayed to the user through the user interface. For example, a countermeasure such as "Check the backup power supply and replace it if necessary" is displayed on the screen. As a result, the user can quickly understand the type and cause of the alarm and take appropriate countermeasures.

[0370] The specific program processing flow is explained below. When a user inputs an alarm message and sends it to the server via the terminal, the server receives the message and saves it as a log. The message is then analyzed by a natural language processing engine, and tokenization and keyword extraction are performed. The extracted keywords are classified using a machine learning model, and appropriate countermeasures are searched for in the database based on the obtained alarm type. The searched countermeasures are finally displayed to the user via the terminal.

[0371] This system solves the problem of alarm messages being difficult to understand, allowing users to respond appropriately and quickly. For example, if the message "Power supply failure detected at base station 5" is input, the system will classify this as a "power supply failure" and suggest the countermeasure of "checking the backup power supply and replacing it if necessary." As a result, users can carry out their work more efficiently.

[0372] The processing flow will be explained below.

[0373] Step 1:

[0374] The user inputs an alarm message into the terminal, for example, "Power supply failure detected at base station 5."

[0375] Step 2:

[0376] The terminal sends the entered alarm message to the server. The message is sent to the server as a character string.

[0377] Step 3:

[0378] The server receives alarm messages sent from the terminals and records the received messages in a database as a log.

[0379] Step 4:

[0380] The server passes the received message to a natural language processing engine (NLP engine), which tokenizes the message and extracts keywords and important phrases.

[0381] For example, keywords such as "Power supply failure" and "base station 5" are extracted.

[0382] Step 5:

[0383] The server inputs the extracted keywords into a machine learning model, which then classifies the alarm type based on these keywords.

[0384] For example, it may be classified as "power supply failure" based on keywords.

[0385] Step 6:

[0386] Based on the alarm classification, the server searches for the appropriate countermeasure from a database that records past response cases and corrective actions.

[0387] Step 7:

[0388] The server converts the detected countermeasure information into a format for display on the user interface. For example, a countermeasure such as "Check the standby power supply and replace it if necessary" is selected.

[0389] Step 8:

[0390] The server generates and sends formatted countermeasure information to the terminal.

[0391] Step 9:

[0392] The terminal displays the countermeasure information received from the server to the user. The user can see information such as "Alarm type: Power supply failure" and "Corrective action: Check the backup power supply and replace it if necessary" on the screen.

[0393] Step 10:

[0394] The user can take appropriate action based on the displayed instructions, thereby quickly implementing measures to resolve the cause of the alarm.

[0395] Example 1

[0396] Next, a description will be given of 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."

[0397] There is a need for a system that can quickly and accurately analyze the contents of alarm messages that occur during base station construction and propose appropriate countermeasures. However, conventional systems have slow processes for understanding alarm messages and proposing countermeasures, making efficient problem solving difficult. For this reason, there is a need to consistently automate the rapid classification of alarms and the proposal of countermeasures, thereby reducing the burden on users.

[0398] 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.

[0399] In this invention, the server includes means for a user to input an alarm message via a terminal, means for receiving the alarm message and saving it as a log, and means for tokenizing the alarm message using a natural language processing engine and extracting keywords, thereby enabling the alarm message to be quickly analyzed and appropriate countermeasures to be presented.

[0400] A "user" is a person or organization that utilizes the system to enter alarm messages.

[0401] A "terminal" is an electronic device that a user uses to input an alarm message and transmit the message to a server.

[0402] An "alarm message" is a message that indicates an abnormality or problem that occurs during the construction of a base station.

[0403] A "server" is a central computer system that receives, analyzes, and processes alarm messages.

[0404] A "log" is a recording file or database that stores received alarm messages.

[0405] A "natural language processing engine" is a software program that analyzes alarm messages and performs tokenization and keyword extraction.

[0406] "Tokenization" is a process of dividing an alarm message into units such as words and phrases.

[0407] "Keywords" are important words or phrases extracted from alarm messages.

[0408] A "machine learning model" is an algorithm for classifying alarm types based on extracted keywords.

[0409] "Alarm type" refers to a category of abnormalities or problems classified based on keywords.

[0410] A "database" is an information collection system that stores past data and knowledge bases and searches for appropriate countermeasures.

[0411] "User interface" refers to the screens and operating means through which a user interacts with a system.

[0412] "Countermeasures" are specific actions or treatments recommended for a classified alarm.

[0413] This invention relates to a system that inputs alarm messages that occur during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures. This system consists of a user interface, a natural language processing engine on the server, a machine learning model, and a database.

[0414] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5.", and the terminal then sends this message to the server.

[0415] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[0416] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[0417] The detected countermeasures are displayed to the user through the user interface. For example, a countermeasure such as "Check the backup power supply and replace it if necessary" is displayed on the screen. As a result, the user can quickly understand the type and cause of the alarm and take appropriate countermeasures.

[0418] For example, if a user inputs the message "Power supply failure detected at base station 5," the system will classify this as a "power supply failure" and suggest a solution: "Check the backup power supply and replace it if necessary." As a result, the user can solve the problem appropriately and quickly.

[0419] Hardware and software used

[0420] The terminals are general-purpose personal computers or smartphones. Secure HTTPS is used as the communication protocol. A high-performance computer server is recommended. Natural language processing engines such as spaCy and NLTK can be used. Machine learning models are built using scikit-learn and TensorFlow. Widely used database management systems such as MySQL and PostgreSQL are used as the database.

[0421] Prompt Sentence Examples

[0422] User Prompt: "Power supply failure detected at base station 5."

[0423] Expected AI Response: "Category: Power Supply Failure. Recommended Action: Check backup power supply and replace if necessary."

[0424] This system allows users to quickly understand the content of alarm messages and take appropriate action. Specifically, if the prompt "Power supply failure detected at base station 5" is entered, the system will classify this as a "power supply failure" and provide the user with appropriate countermeasures. This improves the work efficiency of system users.

[0425] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0426] Step 1:

[0427] The user inputs an alarm message through the terminal. Specifically, the user inputs "Power supply failure detected at base station 5." into the text box on the terminal. When the user sends the input, the terminal sends this message to the server. The input is the alarm message in text format, and the output is the data to be sent to the server.

[0428] Step 2:

[0429] The terminal sends the alarm message to the server. Specifically, the terminal sends the entered message to the server using the secure HTTPS protocol. The input is the text message entered by the user, and the output is the raw alarm message that arrives at the server.

[0430] Step 3:

[0431] The server saves alarm messages in a log. Specifically, the server saves the messages it receives in a database. At this time, the message timestamp and sender information are also saved. The input is the alarm message sent from the terminal, and the output is the log entry saved in the database.

[0432] Step 4:

[0433] The server parses the message using a natural language processing engine. Specifically, the server passes the stored message to a natural language processing engine (e.g., spaCy or NLTK) for tokenization and parsing. The input is the stored alarm message, and the output is a tokenized keyword list.

[0434] Step 5:

[0435] A natural language processing engine tokenizes the message and extracts keywords, such as "Power supply failure" and "base station 5." The input is the text message waiting to be parsed, and the output is the extracted keywords.

[0436] Step 6:

[0437] The server inputs the extracted keywords into a machine learning model. Specifically, the server inputs the keyword list into a machine learning model (e.g., a model built with scikit-learn or TensorFlow) to classify the alarm type. The input is the extracted keywords, and the output is the classified alarm type (e.g., "power supply failure").

[0438] Step 7:

[0439] The machine learning model classifies alarm types, specifically by categorizing keywords into categories such as "power supply failure." The input is a list of keywords, and the output is the alarm type category.

[0440] Step 8:

[0441] The server searches the database for appropriate countermeasures. Specifically, based on the classified alarm type, the server refers to past data and a knowledge base in the database to search for appropriate countermeasures. The input is the classified alarm type, and the output is a list of countermeasures (for example, "Check the backup power supply and replace it if necessary").

[0442] Step 9:

[0443] The server displays the search results to the user through a user interface. Specifically, the countermeasures searched by the server are displayed on the user's device through the user interface. The input is a list of countermeasures, and the output is the displayed countermeasure information. The user can check this and quickly implement countermeasures.

[0444] (Application example 1)

[0445] Next, a description will be given of Application 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."

[0446] When an abnormality occurs in equipment at a manufacturing site, it is necessary to quickly identify the cause of the abnormality and take appropriate measures. However, it is difficult for on-site workers to respond immediately to all abnormalities, which results in a decrease in production efficiency. To solve this problem, a system that automatically detects abnormalities and proposes countermeasures is needed.

[0447] 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.

[0448] In this invention, the server includes means having a user interface for inputting alarm messages, means having a natural language processing engine that receives the alarm messages, tokenizes them, and extracts keywords, means including a machine learning model that classifies alarm types based on the keywords, means for searching a database for appropriate countermeasures for the alarm types and displaying the results on the user interface, means for transmitting alarm messages from devices in the factory that are equipped with anomaly detection sensors, and means for analyzing the alarm messages and suggesting appropriate countermeasures to factory workers. This makes it possible to quickly identify the cause of an anomaly and take appropriate countermeasures.

[0449] An "alarm message" is a message that notifies of an abnormality detected by an abnormality detection sensor in a factory.

[0450] A "user interface" is an interface that allows factory workers to access the system and perform input and output.

[0451] The "natural language processing engine" is an engine that analyzes input alarm messages, tokenizes them, and extracts keywords.

[0452] "Tokenization" is the process of dividing and extracting words and phrases from a text.

[0453] "Keywords" are important words or phrases extracted from the input alarm message.

[0454] A "machine learning model" is an algorithm that learns from data and classifies alarm types.

[0455] A "database" is a system for storing and searching past data and countermeasures.

[0456] An "abnormality detection sensor" is a sensor installed to detect abnormalities in factory equipment.

[0457] "Factory workers" are workers who operate production equipment in factories and perform maintenance and inspections.

[0458] "Countermeasure proposal" is the act of proposing an appropriate solution to a detected abnormality.

[0459] A "server" is a computer system that receives alarm messages and performs analytical processing.

[0460] The present invention relates to a system that detects and analyzes abnormalities that occur in a factory and proposes appropriate countermeasures. This system is composed of a user interface, a natural language processing engine installed on a server, a machine learning model, and a database. A specific embodiment of this system is described below.

[0461] First, when an anomaly detection sensor installed in the factory detects an abnormality, an alarm message is generated. This message might say, for example, "The conveyor belt has stopped." This alarm message is sent from a terminal in the factory to the server.

[0462] The server stores the received alarm messages as logs. These logs contain the data needed for later analysis. A natural language processing engine then analyzes the received messages, tokenizing them and extracting keywords. In this example, keywords such as "conveyor belt" and "stop" are extracted.

[0463] The extracted keywords are input into a machine learning model, which classifies the alarm type. For example, in this case, it is classified as "machine stopped." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base. The countermeasures found are displayed to factory workers through a user interface. For example, a countermeasure such as "Check the motor and replace it if necessary" is displayed on the screen.

[0464] This system enables factory workers to quickly understand the nature and cause of abnormalities and take appropriate measures. Specifically, factory workers can access the system using tablet devices and respond by following the instructions provided by the server.

[0465] The hardware used includes anomaly detection sensors, factory terminals (e.g., tablets), and servers, while the software used includes natural language processing engines (e.g., spaCy and nltk), machine learning models (e.g., scikit-learn and TensorFlow), and databases (e.g., SQLite and MySQL).

[0466] For example, if an alarm message such as "The conveyor belt has stopped" is input, spaCy extracts the keywords "conveyor belt" and "stopped," and scikit-learn classifies it as "machine stopped," and suggests appropriate countermeasures from the database, such as "check the motor and replace it if necessary." In this way, factory workers can immediately take countermeasures.

[0467] An example of a prompt sentence is "The conveyor belt has stopped." By inputting this prompt sentence into the system, the system is designed to suggest the optimal solution.

[0468] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0469] Step 1:

[0470] The abnormality detection sensor detects an abnormality.

[0471] Specific operation: Anomaly detection sensors in the factory detect machine stoppages or abnormal operation.

[0472] Input: Signal from anomaly detection sensor.

[0473] Output: An abnormal message is generated (e.g., "The conveyor belt has stopped.").

[0474] Step 2:

[0475] Send an abnormality message to the server.

[0476] Specific operation: Upon receiving a signal from an anomaly detection sensor, a terminal in the factory generates an alarm message and sends it to a server via the Internet.

[0477] Input: Anomaly message (e.g. "The conveyor belt has stopped.").

[0478] Output: The server receives the abnormal message.

[0479] Step 3:

[0480] The server stores the error message in a log.

[0481] Specific operation: The server records the received abnormal messages in a text file or database.

[0482] Input: Anomaly message (e.g. "The conveyor belt has stopped.").

[0483] Output: Abnormal messages are saved to a log file or database.

[0484] Step 4:

[0485] The server analyzes the abnormal message using a natural language processing engine.

[0486] Specific operation: Tokenize the anomaly message using a natural language processing engine (e.g., spaCy) and extract keywords.

[0487] Input: Anomaly message from a log file or database (e.g., "The conveyor belt has stopped.").

[0488] Output: Extracted keywords (e.g. "conveyor belt", "stop").

[0489] Step 5:

[0490] The server uses a machine learning model to classify the alarm type.

[0491] Specific operation: Using a machine learning model (e.g., scikit-learn), classify the alarm type based on the extracted keywords.

[0492] Input: Extracted keywords (e.g., "conveyor belt", "stop").

[0493] Output: Type of alarm (e.g. "Machine Stop").

[0494] Step 6:

[0495] The server searches the database for an appropriate solution.

[0496] Specific operation: Based on the classified alarm type, the system searches the database for past data and knowledge base to find the appropriate countermeasure.

[0497] Input: Type of alarm (e.g. "Machine Stop").

[0498] Output: Appropriate measures (e.g. "Check motor and replace if necessary").

[0499] Step 7:

[0500] The server displays the appropriate measures on the user interface.

[0501] Specific operation: After finding an appropriate countermeasure, it displays it on the user interface used by factory workers. If necessary, it outputs instructions and procedures.

[0502] Input: Appropriate action (e.g. "Check motor and replace if necessary").

[0503] Output: The countermeasures are displayed in the user interface.

[0504] Furthermore, 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.

[0505] This invention relates to a system that inputs alarm messages generated during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures, and also combines them with an emotion engine that recognizes user emotions. This system is composed of a user interface, a natural language processing engine on a server, a machine learning model, an emotion engine, and a database.

[0506] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5." The terminal then sends this message to the server.

[0507] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[0508] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[0509] Additionally, an emotion engine is also in operation to recognize the emotions the user is feeling when they type. The emotion engine analyzes the emotions from the text the user enters and can identify emotions such as "anxiety" or "tension." Based on this information, the server adjusts the way it presents solutions. For example, if the user is feeling very stressed, it will use calmer, more positive expressions.

[0510] The system then displays advice tailored to the user based on the searched solutions and the emotion engine results. For example, if the solution "Check the backup power supply and replace it if necessary" is found, the system also displays encouraging words such as "Remain calm and continue working. The problem can be solved."

[0511] The specific program processing flow is explained below. When a user inputs an alarm message and sends it to the server via their terminal, the server receives the message and saves it as a log. The message is then analyzed by a natural language processing engine, and tokenization and keyword extraction are performed. The extracted keywords are classified using a machine learning model, and appropriate countermeasures are searched for in a database based on the obtained alarm type. At the same time, the emotion engine analyzes the user's emotions. The way the countermeasures are presented is adjusted based on the emotional information. Finally, countermeasure information and an encouraging message corresponding to the user's emotions are displayed to the user via their terminal.

[0512] This system solves the problem of difficulty in understanding alarm messages and enables appropriate and prompt responses according to the user's emotional state. For example, if the message "Power supply failure detected at base station 5" is input, the system will classify this as a "power supply failure" and provide measures such as "check the backup power supply and replace it if necessary," as well as an encouraging message such as "please stay calm and continue working. The problem can be solved." As a result, the user can carry out their work efficiently and with less mental strain.

[0513] The processing flow will be explained below.

[0514] Step 1:

[0515] The user inputs an alarm message into the terminal, for example, "Power supply failure detected at base station 5."

[0516] Step 2:

[0517] The terminal sends the entered alarm message to the server. The message is sent to the server as a character string.

[0518] Step 3:

[0519] The server receives alarm messages sent from the terminals and records the received messages in a database as a log.

[0520] Step 4:

[0521] The server passes the received message to a natural language processing engine (NLP engine), which tokenizes the message and extracts keywords and important phrases, such as "Power supply failure" and "base station 5."

[0522] Step 5:

[0523] The server inputs the extracted keywords into a machine learning model, which then classifies the alarm type based on these keywords. For example, the alarm may be classified as "power supply failure" based on the keywords.

[0524] Step 6:

[0525] Based on the alarm classification, the server searches for the appropriate countermeasure from a database that records past response cases and corrective actions.

[0526] Step 7:

[0527] To recognize the emotions expressed when a user inputs an alarm message, the server runs an emotion engine. The emotion engine extracts emotion data from the input message and user operations, and analyzes the user's emotional state, such as "anxiety" or "tension."

[0528] Step 8:

[0529] The server analyzes the emotional information recognized by the emotion engine and adjusts the way the countermeasures are presented. If the user is feeling strong stress, it will use calmer and more positive expressions.

[0530] Step 9:

[0531] The server converts the detected countermeasure information and adjusted information based on it into a format for display in the user interface. For example, in addition to the countermeasure "Check the backup power supply and replace it if necessary," it generates an encouraging message such as "Please stay calm and continue working. The problem can be solved."

[0532] Step 10:

[0533] The server sends the generated formatted countermeasure information and the adjusted information to the terminal.

[0534] Step 11:

[0535] The device receives countermeasure information from the server and displays advice based on the user's emotion. The user can see information on the screen such as "Alarm type: Power supply failure," "Corrective action: Check the backup power supply and replace it if necessary," and "Please remain calm and continue working. The problem can be solved."

[0536] Step 12:

[0537] The user can take appropriate action based on the displayed instructions and encouraging messages, thereby quickly implementing measures to resolve the cause of the alarm.

[0538] Example 2

[0539] Next, a description will be given of 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."

[0540] Conventional alarm message analysis systems analyzed alarm messages and proposed appropriate countermeasures, but because they ignored the emotions felt by users when inputting messages, the countermeasures were sometimes not accepted or implemented smoothly. Furthermore, the lack of appropriate support for emotions such as anxiety and tension felt by users led to problems such as a heavy mental burden on users and reduced work efficiency. To solve these problems, a system that not only analyzes alarm messages but also takes into account the user's emotions is needed.

[0541] 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.

[0542] In this invention, the server includes means for saving alarm messages as logs, means having an emotion analysis engine for analyzing emotions from text entered by a user, and means for adjusting the manner in which countermeasures are presented based on the analyzed emotion information. This not only analyzes the contents of alarm messages and proposes appropriate countermeasures, but also adjusts the manner in which the countermeasures are presented according to the emotions felt by the user, thereby reducing mental burden and improving work efficiency.

[0543] An "alarm message" is a message used to report problems or abnormalities that occur on the system.

[0544] "User interface" refers to the screens and input devices that allow a user to interact with a system.

[0545] A "server" is a computer system that processes and stores data over a network.

[0546] A "natural language processing engine" is a technology that analyzes human language and allows computers to understand and analyze it.

[0547] "Tokenization" is the process of dividing a sentence into units such as words and phrases.

[0548] "Keywords" are words or phrases that are particularly important in a piece of text.

[0549] A "machine learning model" is an algorithm that learns from data and makes predictions and classifications.

[0550] A "database" is a system for efficiently managing and searching large amounts of data.

[0551] An "emotion analysis engine" is a technology that analyzes and identifies emotions from text data.

[0552] A "log" is recorded data that a system generates during operation.

[0553] The present invention is a system for analyzing alarm messages related to base station construction and proposing countermeasures, and is configured to propose countermeasures taking into consideration the user's emotions. This system is composed of a user interface, a natural language processing engine installed on a server, a machine learning model, a sentiment analysis engine, and a database.

[0554] First, the user inputs an alarm message through the terminal. For example, if the user inputs the message "Power supply failure detected at base station 5," the terminal sends this message to the server. The server receives the message and stores it in a database as a log. The stored message is then analyzed by a natural language processing engine and tokenized into words and phrases. In this case, keywords such as "Power supply failure" and "base station 5" are extracted.

[0555] The extracted keywords are then fed into a machine learning model, which uses a pre-trained algorithm to classify the alarm type based on the keywords. For example, it may be classified as "power supply failure." The server then searches a database for an appropriate countermeasure for the classified alarm. For example, it may find a countermeasure such as "check the backup power supply and replace it if necessary."

[0556] Furthermore, an emotion analysis engine analyzes the emotions from the text entered by the user. The emotion analysis engine identifies emotions such as "anxiety" or "tension" felt by the user. Based on this emotional information, the way the countermeasures are presented is adjusted. For example, if the user is feeling very anxious, the server will add encouraging words such as "Please stay calm and continue working. The problem can be solved."

[0557] Finally, the adjusted countermeasures and advice are displayed to the user via the terminal. As a specific example, if the user inputs the alarm message "Power supply failure detected at base station 5," the system will classify this as a "power supply failure" and display the countermeasure "Check the backup power supply and replace it if necessary," along with encouraging words such as "Please stay calm and work. The problem can be solved." An example of this prompt would be "Enter the alarm message and output the analysis results and countermeasures."

[0558] This system not only improves analysis accuracy compared to conventional systems, but also takes the user's emotions into consideration, reducing mental burden and enabling more appropriate countermeasures to be provided more quickly.

[0559] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0560] Step 1:

[0561] The user enters an alarm message

[0562] The user inputs an alarm message into the input field of the terminal. For example, the user inputs "Power supply failure detected at base station 5." This input is sent to the terminal as text data.

[0563] Step 2:

[0564] The device sends a message to the server

[0565] The device sends the input message to the server using an HTTP request, which is passed to the server as text data. At this time, the data is encoded into a common data format such as JSON.

[0566] Step 3:

[0567] The server receives the message and stores it in a log.

[0568] The server receives the HTTP request and stores it in temporary memory. Then, it saves the message as a log in a database. A database management system such as MySQL or PostgreSQL is used for saving. The input data is in text format, and the output is saved as a log file or a record in the database.

[0569] Step 4:

[0570] The server analyzes the message using a natural language processing engine

[0571] The server inputs the stored messages into a natural language processing engine. For example, it uses Python's NLTK library to tokenize the messages, which extracts keywords such as "Power supply failure" and "base station 5." The input is text data, and the output is a list of keywords.

[0572] Step 5:

[0573] The server classifies the alarm type using a machine learning model.

[0574] The server inputs the extracted keywords into a pre-trained machine learning model, for example, Scikit-learn, to classify the alarm type. The input is a list of keywords, and the output is an alarm type, such as "power supply failure."

[0575] Step 6:

[0576] The server searches the database for the appropriate solution

[0577] The server searches the database for appropriate countermeasures based on the classified alarm type. The database stores past alarm response history and a knowledge base. The input is the alarm type, and the output is a countermeasure such as "check the standby power supply and replace it if necessary."

[0578] Step 7:

[0579] The server analyzes the user's emotions using an emotion analysis engine.

[0580] The server uses a sentiment analysis engine to analyze the sentiment of the user's input text. For example, it uses Python's TextBlob library to identify emotions such as "anxiety" or "tension" that the user is feeling. The input is text data, and the output is an emotion label.

[0581] Step 8:

[0582] The server adjusts how the measures are expressed.

[0583] The server adjusts the presentation of the countermeasures it provides based on the analyzed emotion information. For example, if the user feels "anxious," it adds encouraging words such as "Please stay calm and work. The problem can be solved." The input is the emotion label and countermeasure data, and the output is the adjusted countermeasure message.

[0584] Step 9:

[0585] The device displays countermeasures and advice to the user

[0586] The terminal displays the countermeasure information sent from the server and the adjusted advice to the user. For example, the terminal displays "Check the backup power supply and replace it if necessary. Please stay calm and work. The problem can be solved." The input is the adjusted countermeasure message, and the output is the display on the user interface.

[0587] (Application example 2)

[0588] Next, a description will be given of Application 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."

[0589] In conventional factory robot operations, it has been difficult to immediately analyze error messages from the robot and provide appropriate countermeasures. Furthermore, if a robot operator feels anxious or stressed by an error message, there is a lack of means to provide support that responds to their emotions. This makes it difficult for operators to work efficiently and with peace of mind, and there is a risk of equipment stalling or incorrect operation.

[0590] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means having a user interface for inputting an alarm message; means equipped with a natural language processing engine that receives the alarm message, tokenizes it, and extracts keywords; means including a machine learning model that classifies the alarm type based on the keywords; means for searching a database for appropriate countermeasures for the alarm type and displaying the results on the user interface; means including an emotion analysis engine that recognizes the emotion of the user when entering the message; and means for adjusting the way advice is presented based on the results of the emotion analysis engine. This makes it possible to quickly and accurately analyze error messages from factory robots and provide appropriate countermeasures, as well as provide appropriate support according to the operator's emotional state.

[0591] Below are definitions of key terms found in the patent claims:

[0592] An "alarm message" refers to a notification sent when a device such as a robot or machine detects an abnormality or error.

[0593] "User interface" refers to the means or screen through which a user directly interacts with a computer system or application.

[0594] A "natural language processing engine" refers to the technology or system that enables a computer to understand and analyze human language and extract meaning.

[0595] "Tokenization" refers to the process of dividing a sentence into words and phrases.

[0596] A "keyword" is a word that has a particularly important meaning in a sentence or paragraph.

[0597] A "machine learning model" refers to an algorithm or system that learns patterns from data and uses those patterns to make inferences and classify new data.

[0598] A "database" refers to a system that systematically stores and manages data, allowing it to be quickly searched and used as needed.

[0599] An "emotion analysis engine" refers to a system that recognizes and analyzes emotions from user input, tone of voice, facial expressions, etc.

[0600] "Adjusting advice presentation" refers to changing the wording or presentation of advice or instructions provided depending on the user's emotional state.

[0601] This invention is a system that allows factory robot operators to analyze error messages from the robot and quickly provide appropriate countermeasures. Furthermore, it can improve work efficiency by recognizing the operator's emotions and providing appropriate support according to those emotions. This system includes the following components:

[0602] 1. Overall system configuration

[0603] The overall system configuration is as follows:

[0604] User interface: A means for operators to input error messages. This can be a factory robot control terminal, a smartphone, or a tablet device.

[0605] Server: A computer system that performs the main processing for analyzing error messages and providing appropriate countermeasures. The server is equipped with a natural language processing engine, machine learning models, a sentiment analysis engine, and a database.

[0606] Natural Language Processing Engine: A library for tokenizing error messages and extracting keywords, for example using Python's TextBlob library.

[0607] Machine learning model: A model for classifying error message types based on extracted keywords. This model is built using the sklearn library.

[0608] Sentiment analysis engine: An engine for recognizing emotions from operator input. Sentiment analysis uses the transformers library and BERT models.

[0609] Database: A source of information that accumulates solutions to error messages. This database stores past error messages and their corresponding solutions.

[0610] 2. System Operation

[0611] The flow for this system to work properly is as follows:

[0612] Alarm message entry and analysis

[0613] When an operator enters an error message on the factory robot's control terminal, the message is sent to the server. The server receives the message and saves it as a log. The natural language processing engine then analyzes the message and extracts the necessary keywords. For example, if the message entered is "Power supply failure detected at manufacturing robot 3," the words "Power supply failure" and "manufacturing robot 3" are extracted.

[0614] Error message classification

[0615] The extracted keywords are input into a machine learning model to classify the type of error message. For example, this might be classified as "power supply failure." For each classified error message, the server accesses a database to search for the appropriate solution.

[0616] Sentiment analysis and advice provision

[0617] The emotion analysis engine analyzes the operator's emotions, identifying emotions such as "anxiety" or "stress" from input such as "This is really stressing me out." Based on this analysis result, the way countermeasures are presented is adjusted. For example, if the operator is feeling anxious, a positive message such as "Please stay calm and continue working. The problem can be solved" is added.

[0618] 3. Specific Examples

[0619] As a specific example, if an operator inputs the alarm message "Power supply failure detected at manufacturing robot 3" and the emotion input "This is really stressing me out," the system will classify this as a "power supply failure" and provide a solution: "Check the backup power supply and replace it if necessary." At the same time, it will also display an encouraging message: "Please stay calm and continue working. The problem can be solved."

[0620] Prompt Sentence Examples

[0621] "I'm experiencing a power supply failure. What's the best way to resolve this issue?"

[0622] "Please advise how to respond if users are feeling uneasy."

[0623] This system enables factory robot operators to deal with error messages quickly and appropriately, while at the same time reducing the psychological burden.

[0624] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0625] Step 1:

[0626] The user enters an error message.

[0627] Specific operation: The user uses the factory robot's management terminal or smartphone to enter an error message such as "Power supply failure detected at manufacturing robot 3."

[0628] Input: Error message

[0629] Output: The entered error message

[0630] Step 2:

[0631] The terminal sends an error message to the server.

[0632] Specific operation: The error message entered by the user is automatically sent from the management terminal to the server.

[0633] Input: Error message

[0634] Output: Error message sent to the server

[0635] Step 3:

[0636] The server receives the error message and stores it in a log.

[0637] Specific behavior: The server records the received error message in a database for later retrieval.

[0638] Input: The error message sent

[0639] Output: Error messages stored in the database

[0640] Step 4:

[0641] The server uses a natural language processing engine to tokenize the error message and extract keywords.

[0642] What it does: It uses Python's TextBlob library to analyze error messages and extract keywords such as "Power supply failure" and "manufacturing robot 3."

[0643] Input: Saved error message

[0644] Output: Extracted keywords

[0645] Step 5:

[0646] The server uses a machine learning model to classify the type of error message based on keywords.

[0647] Specific operation: The extracted keywords are input into a model built with the sklearn library and classified into error message types such as "power supply failure."

[0648] Input: Extracted keywords

[0649] Output: Classified error message types

[0650] Step 6:

[0651] The server searches the database for an appropriate solution based on the type of classified error message.

[0652] Specific operation: For an error message classified as "power supply failure," the database is searched for countermeasures such as "check the backup power supply and replace it if necessary."

[0653] Input: Classified error message type

[0654] Output: Measures found

[0655] Step 7:

[0656] The user inputs emotions.

[0657] Specific operation: The user inputs a feeling such as "This is really stressing me out" into the management terminal.

[0658] Input: Emotion input

[0659] Output: Input emotion text

[0660] Step 8:

[0661] The server analyzes the user's emotions using an emotion analysis engine.

[0662] Specific behavior: Analyzes input emotion text using the BERT model from the transformers library to identify emotions such as "anxiety" and "stress."

[0663] Input: Entered emotion text

[0664] Output: Parsed sentiment

[0665] Step 9:

[0666] The server adjusts the way advice is presented based on the emotions analyzed.

[0667] Specific action: If the analyzed emotion is "anxiety," add an encouraging message to the searched measures: "Please stay calm and work. The problem can be solved."

[0668] Input: Parsed sentiment, retrieved measures

[0669] Output: Adjusted advice

[0670] Step 10:

[0671] The server displays the tailored advice in a user interface.

[0672] Specific operation: Display the adjusted advice on the management terminal screen and provide the user with countermeasures and encouraging messages.

[0673] Input: Tailored Advice

[0674] Output: Advice displayed

[0675] 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 audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio 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 audio data.

[0676] 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> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with 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.

[0677] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0678] [Third embodiment]

[0679] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0680] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0681] 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 WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0682] 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.

[0683] 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 audio in accordance with instructions from the processor 46.

[0684] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and 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 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0685] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0686] Fig. 6 shows an example of the 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.

[0687] 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.

[0688] 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.

[0689] In the headset type terminal 314, a 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.

[0690] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0691] This invention relates to a system that inputs alarm messages that occur during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures. This system consists of a Uzentoface, a natural language processing engine on the server, a machine learning model, and a database.

[0692] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5." The terminal then sends this message to the server.

[0693] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[0694] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[0695] The detected countermeasures are displayed to the user through the user interface. For example, a countermeasure such as "Check the backup power supply and replace it if necessary" is displayed on the screen. As a result, the user can quickly understand the type and cause of the alarm and take appropriate countermeasures.

[0696] The specific program processing flow is explained below. When a user inputs an alarm message and sends it to the server via the terminal, the server receives the message and saves it as a log. The message is then analyzed by a natural language processing engine, and tokenization and keyword extraction are performed. The extracted keywords are classified using a machine learning model, and appropriate countermeasures are searched for in the database based on the obtained alarm type. The searched countermeasures are finally displayed to the user via the terminal.

[0697] This system solves the problem of alarm messages being difficult to understand, allowing users to respond appropriately and quickly. For example, if the message "Power supply failure detected at base station 5" is input, the system will classify this as a "power supply failure" and suggest the countermeasure of "checking the backup power supply and replacing it if necessary." As a result, users can carry out their work more efficiently.

[0698] The processing flow will be explained below.

[0699] Step 1:

[0700] The user inputs an alarm message into the terminal, for example, "Power supply failure detected at base station 5."

[0701] Step 2:

[0702] The terminal sends the entered alarm message to the server. The message is sent to the server as a character string.

[0703] Step 3:

[0704] The server receives alarm messages sent from the terminals and records the received messages in a database as a log.

[0705] Step 4:

[0706] The server passes the received message to a natural language processing engine (NLP engine), which tokenizes the message and extracts keywords and important phrases.

[0707] For example, keywords such as "Power supply failure" and "base station 5" are extracted.

[0708] Step 5:

[0709] The server inputs the extracted keywords into a machine learning model, which then classifies the alarm type based on these keywords.

[0710] For example, it may be classified as "power supply failure" based on keywords.

[0711] Step 6:

[0712] Based on the alarm classification, the server searches for the appropriate countermeasure from a database that records past response cases and corrective actions.

[0713] Step 7:

[0714] The server converts the detected countermeasure information into a format for display on the user interface. For example, a countermeasure such as "Check the standby power supply and replace it if necessary" is selected.

[0715] Step 8:

[0716] The server generates and sends formatted countermeasure information to the terminal.

[0717] Step 9:

[0718] The terminal displays the countermeasure information received from the server to the user. The user can see information such as "Alarm type: Power supply failure" and "Corrective action: Check the backup power supply and replace it if necessary" on the screen.

[0719] Step 10:

[0720] The user can take appropriate action based on the displayed instructions, thereby quickly implementing measures to resolve the cause of the alarm.

[0721] Example 1

[0722] Next, a description will be given of 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."

[0723] There is a need for a system that can quickly and accurately analyze the contents of alarm messages that occur during base station construction and propose appropriate countermeasures. However, conventional systems have slow processes for understanding alarm messages and proposing countermeasures, making efficient problem solving difficult. For this reason, there is a need to consistently automate the rapid classification of alarms and the proposal of countermeasures, thereby reducing the burden on users.

[0724] 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.

[0725] In this invention, the server includes means for a user to input an alarm message via a terminal, means for receiving the alarm message and saving it as a log, and means for tokenizing the alarm message using a natural language processing engine and extracting keywords, thereby enabling the alarm message to be quickly analyzed and appropriate countermeasures to be presented.

[0726] A "user" is a person or organization that utilizes the system to enter alarm messages.

[0727] A "terminal" is an electronic device that a user uses to input an alarm message and transmit the message to a server.

[0728] An "alarm message" is a message that indicates an abnormality or problem that occurs during the construction of a base station.

[0729] A "server" is a central computer system that receives, analyzes, and processes alarm messages.

[0730] A "log" is a recording file or database that stores received alarm messages.

[0731] A "natural language processing engine" is a software program that analyzes alarm messages and performs tokenization and keyword extraction.

[0732] "Tokenization" is a process of dividing an alarm message into units such as words and phrases.

[0733] "Keywords" are important words or phrases extracted from alarm messages.

[0734] A "machine learning model" is an algorithm for classifying alarm types based on extracted keywords.

[0735] "Alarm type" refers to a category of abnormalities or problems classified based on keywords.

[0736] A "database" is an information collection system that stores past data and knowledge bases and searches for appropriate countermeasures.

[0737] "User interface" refers to the screens and operating means through which a user interacts with a system.

[0738] "Countermeasures" are specific actions or treatments recommended for a classified alarm.

[0739] This invention relates to a system that inputs alarm messages that occur during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures. This system consists of a user interface, a natural language processing engine on the server, a machine learning model, and a database.

[0740] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5.", and the terminal then sends this message to the server.

[0741] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[0742] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[0743] The detected countermeasures are displayed to the user through the user interface. For example, a countermeasure such as "Check the backup power supply and replace it if necessary" is displayed on the screen. As a result, the user can quickly understand the type and cause of the alarm and take appropriate countermeasures.

[0744] For example, if a user inputs the message "Power supply failure detected at base station 5," the system will classify this as a "power supply failure" and suggest a solution: "Check the backup power supply and replace it if necessary." As a result, the user can solve the problem appropriately and quickly.

[0745] Hardware and software used

[0746] The terminals are general-purpose personal computers or smartphones. Secure HTTPS is used as the communication protocol. A high-performance computer server is recommended. Natural language processing engines such as spaCy and NLTK can be used. Machine learning models are built using scikit-learn and TensorFlow. Widely used database management systems such as MySQL and PostgreSQL are used as the database.

[0747] Prompt Sentence Examples

[0748] User Prompt: "Power supply failure detected at base station 5."

[0749] Expected AI Response: "Category: Power Supply Failure. Recommended Action: Check backup power supply and replace if necessary."

[0750] This system allows users to quickly understand the content of alarm messages and take appropriate action. Specifically, if the prompt "Power supply failure detected at base station 5" is entered, the system will classify this as a "power supply failure" and provide the user with appropriate countermeasures. This improves the work efficiency of system users.

[0751] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0752] Step 1:

[0753] The user inputs an alarm message through the terminal. Specifically, the user inputs "Power supply failure detected at base station 5." into the text box on the terminal. When the user sends the input, the terminal sends this message to the server. The input is the alarm message in text format, and the output is the data to be sent to the server.

[0754] Step 2:

[0755] The terminal sends the alarm message to the server. Specifically, the terminal sends the entered message to the server using the secure HTTPS protocol. The input is the text message entered by the user, and the output is the raw alarm message that arrives at the server.

[0756] Step 3:

[0757] The server saves alarm messages in a log. Specifically, the server saves the messages it receives in a database. At this time, the message timestamp and sender information are also saved. The input is the alarm message sent from the terminal, and the output is the log entry saved in the database.

[0758] Step 4:

[0759] The server parses the message using a natural language processing engine. Specifically, the server passes the stored message to a natural language processing engine (e.g., spaCy or NLTK) for tokenization and parsing. The input is the stored alarm message, and the output is a tokenized keyword list.

[0760] Step 5:

[0761] A natural language processing engine tokenizes the message and extracts keywords, such as "Power supply failure" and "base station 5." The input is the text message waiting to be parsed, and the output is the extracted keywords.

[0762] Step 6:

[0763] The server inputs the extracted keywords into a machine learning model. Specifically, the server inputs the keyword list into a machine learning model (e.g., a model built with scikit-learn or TensorFlow) to classify the alarm type. The input is the extracted keywords, and the output is the classified alarm type (e.g., "power supply failure").

[0764] Step 7:

[0765] The machine learning model classifies alarm types, specifically by categorizing keywords into categories such as "power supply failure." The input is a list of keywords, and the output is the alarm type category.

[0766] Step 8:

[0767] The server searches the database for appropriate countermeasures. Specifically, based on the classified alarm type, the server refers to past data and a knowledge base in the database to search for appropriate countermeasures. The input is the classified alarm type, and the output is a list of countermeasures (for example, "Check the backup power supply and replace it if necessary").

[0768] Step 9:

[0769] The server displays the search results to the user through a user interface. Specifically, the countermeasures searched by the server are displayed on the user's device through the user interface. The input is a list of countermeasures, and the output is the displayed countermeasure information. The user can check this and quickly implement countermeasures.

[0770] (Application example 1)

[0771] Next, a description will be given of Application 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."

[0772] When an abnormality occurs in equipment at a manufacturing site, it is necessary to quickly identify the cause of the abnormality and take appropriate measures. However, it is difficult for on-site workers to respond immediately to all abnormalities, which results in a decrease in production efficiency. To solve this problem, a system that automatically detects abnormalities and proposes countermeasures is needed.

[0773] 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.

[0774] In this invention, the server includes means having a user interface for inputting alarm messages, means having a natural language processing engine that receives the alarm messages, tokenizes them, and extracts keywords, means including a machine learning model that classifies alarm types based on the keywords, means for searching a database for appropriate countermeasures for the alarm types and displaying the results on the user interface, means for transmitting alarm messages from devices in the factory that are equipped with anomaly detection sensors, and means for analyzing the alarm messages and suggesting appropriate countermeasures to factory workers. This makes it possible to quickly identify the cause of an anomaly and take appropriate countermeasures.

[0775] An "alarm message" is a message that notifies of an abnormality detected by an abnormality detection sensor in a factory.

[0776] A "user interface" is an interface that allows factory workers to access the system and perform input and output.

[0777] The "natural language processing engine" is an engine that analyzes input alarm messages, tokenizes them, and extracts keywords.

[0778] "Tokenization" is the process of dividing and extracting words and phrases from a text.

[0779] "Keywords" are important words or phrases extracted from the input alarm message.

[0780] A "machine learning model" is an algorithm that learns from data and classifies alarm types.

[0781] A "database" is a system for storing and searching past data and countermeasures.

[0782] An "abnormality detection sensor" is a sensor installed to detect abnormalities in factory equipment.

[0783] "Factory workers" are workers who operate production equipment in factories and perform maintenance and inspections.

[0784] "Countermeasure proposal" is the act of proposing an appropriate solution to a detected abnormality.

[0785] A "server" is a computer system that receives alarm messages and performs analytical processing.

[0786] The present invention relates to a system that detects and analyzes abnormalities that occur in a factory and proposes appropriate countermeasures. This system is composed of a user interface, a natural language processing engine installed on a server, a machine learning model, and a database. A specific embodiment of this system is described below.

[0787] First, when an anomaly detection sensor installed in the factory detects an abnormality, an alarm message is generated. This message might say, for example, "The conveyor belt has stopped." This alarm message is sent from a terminal in the factory to the server.

[0788] The server stores the received alarm messages as logs. These logs contain the data needed for later analysis. A natural language processing engine then analyzes the received messages, tokenizing them and extracting keywords. In this example, keywords such as "conveyor belt" and "stop" are extracted.

[0789] The extracted keywords are input into a machine learning model, which classifies the alarm type. For example, in this case, it is classified as "machine stopped." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base. The countermeasures found are displayed to factory workers through a user interface. For example, a countermeasure such as "Check the motor and replace it if necessary" is displayed on the screen.

[0790] This system enables factory workers to quickly understand the nature and cause of abnormalities and take appropriate measures. Specifically, factory workers can access the system using tablet devices and respond by following the instructions provided by the server.

[0791] The hardware used includes anomaly detection sensors, factory terminals (e.g., tablets), and servers, while the software used includes natural language processing engines (e.g., spaCy and nltk), machine learning models (e.g., scikit-learn and TensorFlow), and databases (e.g., SQLite and MySQL).

[0792] For example, if an alarm message such as "The conveyor belt has stopped" is input, spaCy extracts the keywords "conveyor belt" and "stopped," and scikit-learn classifies it as "machine stopped," and suggests appropriate countermeasures from the database, such as "check the motor and replace it if necessary." In this way, factory workers can immediately take countermeasures.

[0793] An example of a prompt sentence is "The conveyor belt has stopped." By inputting this prompt sentence into the system, the system is designed to suggest the optimal solution.

[0794] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0795] Step 1:

[0796] The abnormality detection sensor detects an abnormality.

[0797] Specific operation: Anomaly detection sensors in the factory detect machine stoppages or abnormal operation.

[0798] Input: Signal from anomaly detection sensor.

[0799] Output: An abnormal message is generated (e.g., "The conveyor belt has stopped.").

[0800] Step 2:

[0801] Send an abnormality message to the server.

[0802] Specific operation: Upon receiving a signal from an anomaly detection sensor, a terminal in the factory generates an alarm message and sends it to a server via the Internet.

[0803] Input: Anomaly message (e.g. "The conveyor belt has stopped.").

[0804] Output: The server receives the abnormal message.

[0805] Step 3:

[0806] The server stores the error message in a log.

[0807] Specific operation: The server records the received abnormal messages in a text file or database.

[0808] Input: Anomaly message (e.g. "The conveyor belt has stopped.").

[0809] Output: Abnormal messages are saved to a log file or database.

[0810] Step 4:

[0811] The server analyzes the abnormal message using a natural language processing engine.

[0812] Specific operation: Tokenize the anomaly message using a natural language processing engine (e.g., spaCy) and extract keywords.

[0813] Input: Anomaly message from a log file or database (e.g., "The conveyor belt has stopped.").

[0814] Output: Extracted keywords (e.g. "conveyor belt", "stop").

[0815] Step 5:

[0816] The server uses a machine learning model to classify the alarm type.

[0817] Specific operation: Using a machine learning model (e.g., scikit-learn), classify the alarm type based on the extracted keywords.

[0818] Input: Extracted keywords (e.g., "conveyor belt", "stop").

[0819] Output: Type of alarm (e.g. "Machine Stop").

[0820] Step 6:

[0821] The server searches the database for an appropriate solution.

[0822] Specific operation: Based on the classified alarm type, the system searches the database for past data and knowledge base to find the appropriate countermeasure.

[0823] Input: Type of alarm (e.g. "Machine Stop").

[0824] Output: Appropriate measures (e.g. "Check motor and replace if necessary").

[0825] Step 7:

[0826] The server displays the appropriate measures on the user interface.

[0827] Specific operation: After finding an appropriate countermeasure, it displays it on the user interface used by factory workers. If necessary, it outputs instructions and procedures.

[0828] Input: Appropriate action (e.g. "Check motor and replace if necessary").

[0829] Output: The countermeasures are displayed in the user interface.

[0830] Furthermore, 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.

[0831] This invention relates to a system that inputs alarm messages generated during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures, and also combines them with an emotion engine that recognizes user emotions. This system is composed of a user interface, a natural language processing engine on a server, a machine learning model, an emotion engine, and a database.

[0832] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5." The terminal then sends this message to the server.

[0833] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[0834] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[0835] Additionally, an emotion engine is also in operation to recognize the emotions the user is feeling when they type. The emotion engine analyzes the emotions from the text the user enters and can identify emotions such as "anxiety" or "tension." Based on this information, the server adjusts the way it presents solutions. For example, if the user is feeling very stressed, it will use calmer, more positive expressions.

[0836] The system then displays advice tailored to the user based on the searched solutions and the emotion engine results. For example, if the solution "Check the backup power supply and replace it if necessary" is found, the system also displays encouraging words such as "Remain calm and continue working. The problem can be solved."

[0837] The specific program processing flow is explained below. When a user inputs an alarm message and sends it to the server via their terminal, the server receives the message and saves it as a log. The message is then analyzed by a natural language processing engine, and tokenization and keyword extraction are performed. The extracted keywords are classified using a machine learning model, and appropriate countermeasures are searched for in a database based on the obtained alarm type. At the same time, the emotion engine analyzes the user's emotions. The way the countermeasures are presented is adjusted based on the emotional information. Finally, countermeasure information and an encouraging message corresponding to the user's emotions are displayed to the user via their terminal.

[0838] This system solves the problem of difficulty in understanding alarm messages and enables appropriate and prompt responses according to the user's emotional state. For example, if the message "Power supply failure detected at base station 5" is input, the system will classify this as a "power supply failure" and provide measures such as "check the backup power supply and replace it if necessary," as well as an encouraging message such as "please stay calm and continue working. The problem can be solved." As a result, the user can carry out their work efficiently and with less mental strain.

[0839] The processing flow will be explained below.

[0840] Step 1:

[0841] The user inputs an alarm message into the terminal, for example, "Power supply failure detected at base station 5."

[0842] Step 2:

[0843] The terminal sends the entered alarm message to the server. The message is sent to the server as a character string.

[0844] Step 3:

[0845] The server receives alarm messages sent from the terminals and records the received messages in a database as a log.

[0846] Step 4:

[0847] The server passes the received message to a natural language processing engine (NLP engine), which tokenizes the message and extracts keywords and important phrases, such as "Power supply failure" and "base station 5."

[0848] Step 5:

[0849] The server inputs the extracted keywords into a machine learning model, which then classifies the alarm type based on these keywords. For example, the alarm may be classified as "power supply failure" based on the keywords.

[0850] Step 6:

[0851] Based on the alarm classification, the server searches for the appropriate countermeasure from a database that records past response cases and corrective actions.

[0852] Step 7:

[0853] To recognize the emotions expressed when a user inputs an alarm message, the server runs an emotion engine. The emotion engine extracts emotion data from the input message and user operations, and analyzes the user's emotional state, such as "anxiety" or "tension."

[0854] Step 8:

[0855] The server analyzes the emotional information recognized by the emotion engine and adjusts the way the countermeasures are presented. If the user is feeling strong stress, it will use calmer and more positive expressions.

[0856] Step 9:

[0857] The server converts the detected countermeasure information and adjusted information based on it into a format for display in the user interface. For example, in addition to the countermeasure "Check the backup power supply and replace it if necessary," it generates an encouraging message such as "Please stay calm and continue working. The problem can be solved."

[0858] Step 10:

[0859] The server sends the generated formatted countermeasure information and the adjusted information to the terminal.

[0860] Step 11:

[0861] The device receives countermeasure information from the server and displays advice based on the user's emotion. The user can see information on the screen such as "Alarm type: Power supply failure," "Corrective action: Check the backup power supply and replace it if necessary," and "Please remain calm and continue working. The problem can be solved."

[0862] Step 12:

[0863] The user can take appropriate action based on the displayed instructions and encouraging messages, thereby quickly implementing measures to resolve the cause of the alarm.

[0864] Example 2

[0865] Next, a description will be given of 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."

[0866] Conventional alarm message analysis systems analyzed alarm messages and proposed appropriate countermeasures, but because they ignored the emotions felt by users when inputting messages, the countermeasures were sometimes not accepted or implemented smoothly. Furthermore, the lack of appropriate support for emotions such as anxiety and tension felt by users led to problems such as a heavy mental burden on users and reduced work efficiency. To solve these problems, a system that not only analyzes alarm messages but also takes into account the user's emotions is needed.

[0867] 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.

[0868] In this invention, the server includes means for saving alarm messages as logs, means having an emotion analysis engine for analyzing emotions from text entered by a user, and means for adjusting the manner in which countermeasures are presented based on the analyzed emotion information. This not only analyzes the contents of alarm messages and proposes appropriate countermeasures, but also adjusts the manner in which the countermeasures are presented according to the emotions felt by the user, thereby reducing mental burden and improving work efficiency.

[0869] An "alarm message" is a message used to report problems or abnormalities that occur on the system.

[0870] "User interface" refers to the screens and input devices that allow a user to interact with a system.

[0871] A "server" is a computer system that processes and stores data over a network.

[0872] A "natural language processing engine" is a technology that analyzes human language and allows computers to understand and analyze it.

[0873] "Tokenization" is the process of dividing a sentence into units such as words and phrases.

[0874] "Keywords" are words or phrases that are particularly important in a piece of text.

[0875] A "machine learning model" is an algorithm that learns from data and makes predictions and classifications.

[0876] A "database" is a system for efficiently managing and searching large amounts of data.

[0877] An "emotion analysis engine" is a technology that analyzes and identifies emotions from text data.

[0878] A "log" is recorded data that a system generates during operation.

[0879] The present invention is a system for analyzing alarm messages related to base station construction and proposing countermeasures, and is configured to propose countermeasures taking into consideration the user's emotions. This system is composed of a user interface, a natural language processing engine installed on a server, a machine learning model, a sentiment analysis engine, and a database.

[0880] First, the user inputs an alarm message through the terminal. For example, if the user inputs the message "Power supply failure detected at base station 5," the terminal sends this message to the server. The server receives the message and stores it in a database as a log. The stored message is then analyzed by a natural language processing engine and tokenized into words and phrases. In this case, keywords such as "Power supply failure" and "base station 5" are extracted.

[0881] The extracted keywords are then fed into a machine learning model, which uses a pre-trained algorithm to classify the alarm type based on the keywords. For example, it may be classified as "power supply failure." The server then searches a database for an appropriate countermeasure for the classified alarm. For example, it may find a countermeasure such as "check the backup power supply and replace it if necessary."

[0882] Furthermore, an emotion analysis engine analyzes the emotions from the text entered by the user. The emotion analysis engine identifies emotions such as "anxiety" or "tension" felt by the user. Based on this emotional information, the way the countermeasures are presented is adjusted. For example, if the user is feeling very anxious, the server will add encouraging words such as "Please stay calm and continue working. The problem can be solved."

[0883] Finally, the adjusted countermeasures and advice are displayed to the user via the terminal. As a specific example, if the user inputs the alarm message "Power supply failure detected at base station 5," the system will classify this as a "power supply failure" and display the countermeasure "Check the backup power supply and replace it if necessary," along with encouraging words such as "Please stay calm and work. The problem can be solved." An example of this prompt would be "Enter the alarm message and output the analysis results and countermeasures."

[0884] This system not only improves analysis accuracy compared to conventional systems, but also takes the user's emotions into consideration, reducing mental burden and enabling more appropriate countermeasures to be provided more quickly.

[0885] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0886] Step 1:

[0887] The user enters an alarm message

[0888] The user inputs an alarm message into the input field of the terminal. For example, the user inputs "Power supply failure detected at base station 5." This input is sent to the terminal as text data.

[0889] Step 2:

[0890] The device sends a message to the server

[0891] The device sends the input message to the server using an HTTP request, which is passed to the server as text data. At this time, the data is encoded into a common data format such as JSON.

[0892] Step 3:

[0893] The server receives the message and stores it in a log.

[0894] The server receives the HTTP request and stores it in temporary memory. Then, it saves the message as a log in a database. A database management system such as MySQL or PostgreSQL is used for saving. The input data is in text format, and the output is saved as a log file or a record in the database.

[0895] Step 4:

[0896] The server analyzes the message using a natural language processing engine

[0897] The server inputs the stored messages into a natural language processing engine. For example, it uses Python's NLTK library to tokenize the messages, which extracts keywords such as "Power supply failure" and "base station 5." The input is text data, and the output is a list of keywords.

[0898] Step 5:

[0899] The server classifies the alarm type using a machine learning model.

[0900] The server inputs the extracted keywords into a pre-trained machine learning model, for example, Scikit-learn, to classify the alarm type. The input is a list of keywords, and the output is an alarm type, such as "power supply failure."

[0901] Step 6:

[0902] The server searches the database for the appropriate solution

[0903] The server searches the database for appropriate countermeasures based on the classified alarm type. The database stores past alarm response history and a knowledge base. The input is the alarm type, and the output is a countermeasure such as "check the standby power supply and replace it if necessary."

[0904] Step 7:

[0905] The server analyzes the user's emotions using an emotion analysis engine.

[0906] The server uses a sentiment analysis engine to analyze the sentiment of the user's input text. For example, it uses Python's TextBlob library to identify emotions such as "anxiety" or "tension" that the user is feeling. The input is text data, and the output is an emotion label.

[0907] Step 8:

[0908] The server adjusts how the measures are expressed.

[0909] The server adjusts the presentation of the countermeasures it provides based on the analyzed emotion information. For example, if the user feels "anxious," it adds encouraging words such as "Please stay calm and work. The problem can be solved." The input is the emotion label and countermeasure data, and the output is the adjusted countermeasure message.

[0910] Step 9:

[0911] The device displays countermeasures and advice to the user

[0912] The terminal displays the countermeasure information sent from the server and the adjusted advice to the user. For example, the terminal displays "Check the backup power supply and replace it if necessary. Please stay calm and work. The problem can be solved." The input is the adjusted countermeasure message, and the output is the display on the user interface.

[0913] (Application example 2)

[0914] Next, a description will be given of Application 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."

[0915] In conventional factory robot operations, it has been difficult to immediately analyze error messages from the robot and provide appropriate countermeasures. Furthermore, if a robot operator feels anxious or stressed by an error message, there is a lack of means to provide support that responds to their emotions. This makes it difficult for operators to work efficiently and with peace of mind, and there is a risk of equipment stalling or incorrect operation.

[0916] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means having a user interface for inputting an alarm message; means equipped with a natural language processing engine that receives the alarm message, tokenizes it, and extracts keywords; means including a machine learning model that classifies the alarm type based on the keywords; means for searching a database for appropriate countermeasures for the alarm type and displaying the results on the user interface; means including an emotion analysis engine that recognizes the emotion of the user when entering the message; and means for adjusting the way advice is presented based on the results of the emotion analysis engine. This makes it possible to quickly and accurately analyze error messages from factory robots and provide appropriate countermeasures, as well as provide appropriate support according to the operator's emotional state.

[0917] Below are definitions of key terms found in the patent claims:

[0918] An "alarm message" refers to a notification sent when a device such as a robot or machine detects an abnormality or error.

[0919] "User interface" refers to the means or screen through which a user directly interacts with a computer system or application.

[0920] A "natural language processing engine" refers to the technology or system that enables a computer to understand and analyze human language and extract meaning.

[0921] "Tokenization" refers to the process of dividing a sentence into words and phrases.

[0922] A "keyword" is a word that has a particularly important meaning in a sentence or paragraph.

[0923] A "machine learning model" refers to an algorithm or system that learns patterns from data and uses those patterns to make inferences and classify new data.

[0924] A "database" refers to a system that systematically stores and manages data, allowing it to be quickly searched and used as needed.

[0925] An "emotion analysis engine" refers to a system that recognizes and analyzes emotions from user input, tone of voice, facial expressions, etc.

[0926] "Adjusting advice presentation" refers to changing the wording or presentation of advice or instructions provided depending on the user's emotional state.

[0927] This invention is a system that allows factory robot operators to analyze error messages from the robot and quickly provide appropriate countermeasures. Furthermore, it can improve work efficiency by recognizing the operator's emotions and providing appropriate support according to those emotions. This system includes the following components:

[0928] 1. Overall system configuration

[0929] The overall system configuration is as follows:

[0930] User interface: A means for operators to input error messages. This can be a factory robot control terminal, a smartphone, or a tablet device.

[0931] Server: A computer system that performs the main processing for analyzing error messages and providing appropriate countermeasures. The server is equipped with a natural language processing engine, machine learning models, a sentiment analysis engine, and a database.

[0932] Natural Language Processing Engine: A library for tokenizing error messages and extracting keywords, for example using Python's TextBlob library.

[0933] Machine learning model: A model for classifying error message types based on extracted keywords. This model is built using the sklearn library.

[0934] Sentiment analysis engine: An engine for recognizing emotions from operator input. For sentiment analysis, the transformers library is used, along with the BERT model.

[0935] Database: A source of information that accumulates solutions to error messages. This database stores past error messages and their corresponding solutions.

[0936] 2. System Operation

[0937] The flow for this system to work properly is as follows:

[0938] Alarm message entry and analysis

[0939] When an operator enters an error message on the factory robot's control terminal, the message is sent to the server. The server receives the message and saves it as a log. The natural language processing engine then analyzes the message and extracts the necessary keywords. For example, if the message entered is "Power supply failure detected at manufacturing robot 3," the words "Power supply failure" and "manufacturing robot 3" are extracted.

[0940] Error message classification

[0941] The extracted keywords are input into a machine learning model to classify the type of error message. For example, this one might be classified as "power supply failure." For each classified error message, the server accesses a database to search for the appropriate solution.

[0942] Sentiment analysis and advice provision

[0943] The emotion analysis engine analyzes the operator's emotions, identifying emotions such as "anxiety" or "stress" from input such as "This is really stressing me out." Based on this analysis result, the way countermeasures are presented is adjusted. For example, if the operator is feeling anxious, a positive message such as "Please stay calm and continue working. The problem can be solved" is added.

[0944] 3. Specific Examples

[0945] As a specific example, if an operator inputs the alarm message "Power supply failure detected at manufacturing robot 3" and the emotion input "This is really stressing me out," the system will classify this as a "power supply failure" and provide a solution: "Check the backup power supply and replace it if necessary." At the same time, it will also display an encouraging message: "Please stay calm and continue working. The problem can be solved."

[0946] Prompt Sentence Examples

[0947] "I'm experiencing a power supply failure. What's the best way to resolve this issue?"

[0948] "Please advise how to respond if users are feeling uneasy."

[0949] This system enables factory robot operators to deal with error messages quickly and appropriately, while at the same time reducing the psychological burden.

[0950] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0951] Step 1:

[0952] The user enters an error message.

[0953] Specific operation: The user uses the factory robot's management terminal or smartphone to enter an error message such as "Power supply failure detected at manufacturing robot 3."

[0954] Input: Error message

[0955] Output: The entered error message

[0956] Step 2:

[0957] The terminal sends an error message to the server.

[0958] Specific operation: The error message entered by the user is automatically sent from the management terminal to the server.

[0959] Input: Error message

[0960] Output: Error message sent to the server

[0961] Step 3:

[0962] The server receives the error message and stores it in a log.

[0963] Specific behavior: The server records the received error message in a database for later retrieval.

[0964] Input: The error message sent

[0965] Output: Error messages stored in the database

[0966] Step 4:

[0967] The server uses a natural language processing engine to tokenize the error message and extract keywords.

[0968] What it does: It uses Python's TextBlob library to analyze error messages and extract keywords such as "Power supply failure" and "manufacturing robot 3."

[0969] Input: Saved error message

[0970] Output: Extracted keywords

[0971] Step 5:

[0972] The server uses a machine learning model to classify the type of error message based on keywords.

[0973] Specific operation: The extracted keywords are input into a model built with the sklearn library and classified into error message types such as "power supply failure."

[0974] Input: Extracted keywords

[0975] Output: Classified error message types

[0976] Step 6:

[0977] The server searches the database for an appropriate solution based on the type of classified error message.

[0978] Specific operation: For an error message classified as "power supply failure," the database is searched for countermeasures such as "check the backup power supply and replace it if necessary."

[0979] Input: Classified error message type

[0980] Output: Measures found

[0981] Step 7:

[0982] The user inputs emotions.

[0983] Specific operation: The user inputs a feeling such as "This is really stressing me out" into the management terminal.

[0984] Input: Emotion input

[0985] Output: Input emotion text

[0986] Step 8:

[0987] The server analyzes the user's emotions using an emotion analysis engine.

[0988] Specific behavior: Analyzes input emotion text using the BERT model from the transformers library to identify emotions such as "anxiety" and "stress."

[0989] Input: Entered emotion text

[0990] Output: Parsed sentiment

[0991] Step 9:

[0992] The server adjusts the way advice is presented based on the emotions analyzed.

[0993] Specific action: If the analyzed emotion is "anxiety," add an encouraging message to the searched measures: "Please stay calm and work. The problem can be solved."

[0994] Input: Parsed sentiment, retrieved measures

[0995] Output: Adjusted advice

[0996] Step 10:

[0997] The server displays the tailored advice in a user interface.

[0998] Specific operation: Display the adjusted advice on the management terminal screen and provide the user with countermeasures and encouraging messages.

[0999] Input: Tailored Advice

[1000] Output: Advice displayed

[1001] 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 audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio 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 audio data.

[1002] 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> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with 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.

[1003] 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.

[1004] [Fourth embodiment]

[1005] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1006] 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.

[1007] 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 WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1008] 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. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1009] 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 audio in accordance with instructions from the processor 46.

[1010] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and 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 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1011] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1012] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. 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.

[1013] Fig. 8 shows an example of the 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.

[1014] 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.

[1015] 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.

[1016] In the robot 414, the processor 46 performs the reception output process. 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.

[1017] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. 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."

[1018] This invention relates to a system that inputs alarm messages that occur during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures. This system consists of a Uzentoface, a natural language processing engine on the server, a machine learning model, and a database.

[1019] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5." The terminal then sends this message to the server.

[1020] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[1021] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[1022] The detected countermeasures are displayed to the user through the user interface. For example, a countermeasure such as "Check the backup power supply and replace it if necessary" is displayed on the screen. As a result, the user can quickly understand the type and cause of the alarm and take appropriate countermeasures.

[1023] The specific program processing flow is explained below. When a user inputs an alarm message and sends it to the server via the terminal, the server receives the message and saves it as a log. The message is then analyzed by a natural language processing engine, and tokenization and keyword extraction are performed. The extracted keywords are classified using a machine learning model, and appropriate countermeasures are searched for in the database based on the obtained alarm type. The searched countermeasures are finally displayed to the user via the terminal.

[1024] This system solves the problem of alarm messages being difficult to understand, allowing users to respond appropriately and quickly. For example, if the message "Power supply failure detected at base station 5" is input, the system will classify this as a "power supply failure" and suggest the countermeasure of "checking the backup power supply and replacing it if necessary." As a result, users can carry out their work more efficiently.

[1025] The processing flow will be explained below.

[1026] Step 1:

[1027] The user inputs an alarm message into the terminal, for example, "Power supply failure detected at base station 5."

[1028] Step 2:

[1029] The terminal sends the entered alarm message to the server. The message is sent to the server as a character string.

[1030] Step 3:

[1031] The server receives alarm messages sent from the terminals and records the received messages in a database as a log.

[1032] Step 4:

[1033] The server passes the received message to a natural language processing engine (NLP engine), which tokenizes the message and extracts keywords and important phrases.

[1034] For example, keywords such as "Power supply failure" and "base station 5" are extracted.

[1035] Step 5:

[1036] The server inputs the extracted keywords into a machine learning model, which then classifies the alarm type based on these keywords.

[1037] For example, it may be classified as "power supply failure" based on keywords.

[1038] Step 6:

[1039] Based on the alarm classification, the server searches for the appropriate countermeasure from a database that records past response cases and corrective actions.

[1040] Step 7:

[1041] The server converts the detected countermeasure information into a format for display on the user interface. For example, a countermeasure such as "Check the standby power supply and replace it if necessary" is selected.

[1042] Step 8:

[1043] The server generates and sends formatted countermeasure information to the terminal.

[1044] Step 9:

[1045] The terminal displays the countermeasure information received from the server to the user. The user can see information such as "Alarm type: Power supply failure" and "Corrective action: Check the backup power supply and replace it if necessary" on the screen.

[1046] Step 10:

[1047] The user can take appropriate action based on the displayed instructions, thereby quickly implementing measures to resolve the cause of the alarm.

[1048] Example 1

[1049] Next, a description will be given of 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."

[1050] There is a need for a system that can quickly and accurately analyze the contents of alarm messages that occur during base station construction and propose appropriate countermeasures. However, conventional systems have slow processes for understanding alarm messages and proposing countermeasures, making efficient problem solving difficult. For this reason, there is a need to consistently automate the rapid classification of alarms and the proposal of countermeasures, thereby reducing the burden on users.

[1051] 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.

[1052] In this invention, the server includes means for a user to input an alarm message via a terminal, means for receiving the alarm message and saving it as a log, and means for tokenizing the alarm message using a natural language processing engine and extracting keywords, thereby enabling the alarm message to be quickly analyzed and appropriate countermeasures to be presented.

[1053] A "user" is a person or organization that utilizes the system to enter alarm messages.

[1054] A "terminal" is an electronic device that a user uses to input an alarm message and transmit the message to a server.

[1055] An "alarm message" is a message that indicates an abnormality or problem that occurs during the construction of a base station.

[1056] A "server" is a central computer system that receives, analyzes, and processes alarm messages.

[1057] A "log" is a recording file or database that stores received alarm messages.

[1058] A "natural language processing engine" is a software program that analyzes alarm messages and performs tokenization and keyword extraction.

[1059] "Tokenization" is a process of dividing an alarm message into units such as words and phrases.

[1060] "Keywords" are important words or phrases extracted from alarm messages.

[1061] A "machine learning model" is an algorithm for classifying alarm types based on extracted keywords.

[1062] "Alarm type" refers to a category of abnormalities or problems classified based on keywords.

[1063] A "database" is an information collection system that stores past data and knowledge bases and searches for appropriate countermeasures.

[1064] "User interface" refers to the screens and operating means through which a user interacts with a system.

[1065] "Countermeasures" are specific actions or treatments recommended for a classified alarm.

[1066] This invention relates to a system that inputs alarm messages that occur during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures. This system consists of a user interface, a natural language processing engine on the server, a machine learning model, and a database.

[1067] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5.", and the terminal then sends this message to the server.

[1068] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[1069] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[1070] The detected countermeasures are displayed to the user through the user interface. For example, a countermeasure such as "Check the backup power supply and replace it if necessary" is displayed on the screen. As a result, the user can quickly understand the type and cause of the alarm and take appropriate countermeasures.

[1071] For example, if a user inputs the message "Power supply failure detected at base station 5," the system will classify this as a "power supply failure" and suggest a solution: "Check the backup power supply and replace it if necessary." As a result, the user can solve the problem appropriately and quickly.

[1072] Hardware and software used

[1073] The terminals are general-purpose personal computers or smartphones. Secure HTTPS is used as the communication protocol. A high-performance computer server is recommended. Natural language processing engines such as spaCy and NLTK can be used. Machine learning models are built using scikit-learn and TensorFlow. Widely used database management systems such as MySQL and PostgreSQL are used as the database.

[1074] Prompt Sentence Examples

[1075] User Prompt: "Power supply failure detected at base station 5."

[1076] Expected AI Response: "Category: Power Supply Failure. Recommended Action: Check backup power supply and replace if necessary."

[1077] This system allows users to quickly understand the content of alarm messages and take appropriate action. Specifically, if the prompt "Power supply failure detected at base station 5" is entered, the system will classify this as a "power supply failure" and provide the user with appropriate countermeasures. This improves the work efficiency of system users.

[1078] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1079] Step 1:

[1080] The user inputs an alarm message through the terminal. Specifically, the user inputs "Power supply failure detected at base station 5." into the text box on the terminal. When the user sends the input, the terminal sends this message to the server. The input is the alarm message in text format, and the output is the data to be sent to the server.

[1081] Step 2:

[1082] The terminal sends the alarm message to the server. Specifically, the terminal sends the entered message to the server using the secure HTTPS protocol. The input is the text message entered by the user, and the output is the raw alarm message that arrives at the server.

[1083] Step 3:

[1084] The server saves alarm messages in a log. Specifically, the server saves the messages it receives in a database. At this time, the message timestamp and sender information are also saved. The input is the alarm message sent from the terminal, and the output is the log entry saved in the database.

[1085] Step 4:

[1086] The server parses the message using a natural language processing engine. Specifically, the server passes the stored message to a natural language processing engine (e.g., spaCy or NLTK) for tokenization and parsing. The input is the stored alarm message, and the output is a tokenized keyword list.

[1087] Step 5:

[1088] A natural language processing engine tokenizes the message and extracts keywords, such as "Power supply failure" and "base station 5." The input is the text message waiting to be parsed, and the output is the extracted keywords.

[1089] Step 6:

[1090] The server inputs the extracted keywords into a machine learning model. Specifically, the server inputs the keyword list into a machine learning model (e.g., a model built with scikit-learn or TensorFlow) to classify the alarm type. The input is the extracted keywords, and the output is the classified alarm type (e.g., "power supply failure").

[1091] Step 7:

[1092] The machine learning model classifies alarm types, specifically by categorizing keywords into categories such as "power supply failure." The input is a list of keywords, and the output is the alarm type category.

[1093] Step 8:

[1094] The server searches the database for appropriate countermeasures. Specifically, based on the classified alarm type, the server refers to past data and a knowledge base in the database to search for appropriate countermeasures. The input is the classified alarm type, and the output is a list of countermeasures (for example, "Check the backup power supply and replace it if necessary").

[1095] Step 9:

[1096] The server displays the search results to the user through a user interface. Specifically, the countermeasures searched by the server are displayed on the user's device through the user interface. The input is a list of countermeasures, and the output is the displayed countermeasure information. The user can check this and quickly implement countermeasures.

[1097] (Application example 1)

[1098] Next, a description will be given of Application 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."

[1099] When an abnormality occurs in equipment at a manufacturing site, it is necessary to quickly identify the cause of the abnormality and take appropriate measures. However, it is difficult for on-site workers to respond immediately to all abnormalities, which results in a decrease in production efficiency. To solve this problem, a system that automatically detects abnormalities and proposes countermeasures is needed.

[1100] 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.

[1101] In this invention, the server includes means having a user interface for inputting alarm messages, means having a natural language processing engine that receives the alarm messages, tokenizes them, and extracts keywords, means including a machine learning model that classifies alarm types based on the keywords, means for searching a database for appropriate countermeasures for the alarm types and displaying the results on the user interface, means for transmitting alarm messages from devices in the factory that are equipped with anomaly detection sensors, and means for analyzing the alarm messages and suggesting appropriate countermeasures to factory workers. This makes it possible to quickly identify the cause of an anomaly and take appropriate countermeasures.

[1102] An "alarm message" is a message that notifies of an abnormality detected by an abnormality detection sensor in a factory.

[1103] A "user interface" is an interface that allows factory workers to access the system and perform input and output.

[1104] The "natural language processing engine" is an engine that analyzes input alarm messages, tokenizes them, and extracts keywords.

[1105] "Tokenization" is the process of dividing and extracting words and phrases from a text.

[1106] "Keywords" are important words or phrases extracted from the input alarm message.

[1107] A "machine learning model" is an algorithm that learns from data and classifies alarm types.

[1108] A "database" is a system for storing and searching past data and countermeasures.

[1109] An "abnormality detection sensor" is a sensor installed to detect abnormalities in factory equipment.

[1110] "Factory workers" are workers who operate production equipment in factories and perform maintenance and inspections.

[1111] "Countermeasure proposal" is the act of proposing an appropriate solution to a detected abnormality.

[1112] A "server" is a computer system that receives alarm messages and performs analytical processing.

[1113] The present invention relates to a system that detects and analyzes abnormalities that occur in a factory and proposes appropriate countermeasures. This system is composed of a user interface, a natural language processing engine installed on a server, a machine learning model, and a database. A specific embodiment of this system is described below.

[1114] First, when an anomaly detection sensor installed in the factory detects an abnormality, an alarm message is generated. This message might say, for example, "The conveyor belt has stopped." This alarm message is sent from a terminal in the factory to the server.

[1115] The server stores the received alarm messages as logs. These logs contain the data needed for later analysis. A natural language processing engine then analyzes the received messages, tokenizing them and extracting keywords. In this example, keywords such as "conveyor belt" and "stop" are extracted.

[1116] The extracted keywords are input into a machine learning model, which classifies the alarm type. For example, in this case, it is classified as "machine stopped." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base. The countermeasures found are displayed to factory workers through a user interface. For example, a countermeasure such as "Check the motor and replace it if necessary" is displayed on the screen.

[1117] This system enables factory workers to quickly understand the nature and cause of abnormalities and take appropriate measures. Specifically, factory workers can access the system using tablet devices and respond by following the instructions provided by the server.

[1118] The hardware used includes anomaly detection sensors, factory terminals (e.g., tablets), and servers, while the software used includes natural language processing engines (e.g., spaCy and nltk), machine learning models (e.g., scikit-learn and TensorFlow), and databases (e.g., SQLite and MySQL).

[1119] For example, if an alarm message such as "The conveyor belt has stopped" is input, spaCy extracts the keywords "conveyor belt" and "stopped," and scikit-learn classifies it as "machine stopped," and suggests appropriate countermeasures from the database, such as "check the motor and replace it if necessary." In this way, factory workers can immediately take countermeasures.

[1120] An example of a prompt sentence is "The conveyor belt has stopped." By inputting this prompt sentence into the system, the system is designed to suggest the optimal solution.

[1121] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1122] Step 1:

[1123] The abnormality detection sensor detects an abnormality.

[1124] Specific operation: Anomaly detection sensors in the factory detect machine stoppages or abnormal operation.

[1125] Input: Signal from anomaly detection sensor.

[1126] Output: An abnormal message is generated (e.g., "The conveyor belt has stopped.").

[1127] Step 2:

[1128] Send an abnormality message to the server.

[1129] Specific operation: Upon receiving a signal from an anomaly detection sensor, a terminal in the factory generates an alarm message and sends it to a server via the Internet.

[1130] Input: Anomaly message (e.g. "The conveyor belt has stopped.").

[1131] Output: The server receives the abnormal message.

[1132] Step 3:

[1133] The server stores the error message in a log.

[1134] Specific operation: The server records the received abnormal messages in a text file or database.

[1135] Input: Anomaly message (e.g. "The conveyor belt has stopped.").

[1136] Output: Abnormal messages are saved to a log file or database.

[1137] Step 4:

[1138] The server analyzes the abnormal message using a natural language processing engine.

[1139] Specific operation: Tokenize the anomaly message using a natural language processing engine (e.g., spaCy) and extract keywords.

[1140] Input: Anomaly message from a log file or database (e.g., "The conveyor belt has stopped.").

[1141] Output: Extracted keywords (e.g. "conveyor belt", "stop").

[1142] Step 5:

[1143] The server uses a machine learning model to classify the alarm type.

[1144] Specific operation: Using a machine learning model (e.g., scikit-learn), classify the alarm type based on the extracted keywords.

[1145] Input: Extracted keywords (e.g., "conveyor belt", "stop").

[1146] Output: Type of alarm (e.g. "Machine Stop").

[1147] Step 6:

[1148] The server searches the database for an appropriate solution.

[1149] Specific operation: Based on the classified alarm type, the system searches the database for past data and knowledge base to find the appropriate countermeasure.

[1150] Input: Type of alarm (e.g. "Machine Stop").

[1151] Output: Appropriate measures (e.g. "Check motor and replace if necessary").

[1152] Step 7:

[1153] The server displays the appropriate measures on the user interface.

[1154] Specific operation: After finding an appropriate countermeasure, it displays it on the user interface used by factory workers. If necessary, it outputs instructions and procedures.

[1155] Input: Appropriate action (e.g. "Check motor and replace if necessary").

[1156] Output: The countermeasures are displayed in the user interface.

[1157] Furthermore, 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.

[1158] This invention relates to a system that inputs alarm messages generated during base station construction, analyzes them, explains their contents, and proposes appropriate countermeasures, and also combines them with an emotion engine that recognizes user emotions. This system is composed of a user interface, a natural language processing engine on a server, a machine learning model, an emotion engine, and a database.

[1159] First, a user inputs an alarm message through the terminal, for example, "Power supply failure detected at base station 5." The terminal then sends this message to the server.

[1160] The server stores the received alarm messages as logs. Next, a natural language processing engine analyzes the received messages, tokenizes them, and extracts keywords, such as "Power supply failure" and "base station 5."

[1161] The extracted keywords are input into a machine learning model to classify the alarm type. For example, it may be classified as "power supply failure." For each classified alarm, the server accesses the database and searches for appropriate countermeasures from past data and a knowledge base.

[1162] Additionally, an emotion engine is also in operation to recognize the emotions the user is feeling when they type. The emotion engine analyzes the emotions from the text the user enters and can identify emotions such as "anxiety" or "tension." Based on this information, the server adjusts the way it presents solutions. For example, if the user is feeling very stressed, it will use calmer, more positive expressions.

[1163] The system then displays advice tailored to the user based on the searched solutions and the emotion engine results. For example, if the solution "Check the backup power supply and replace it if necessary" is found, the system also displays encouraging words such as "Remain calm and continue working. The problem can be solved."

[1164] The specific program processing flow is explained below. When a user inputs an alarm message and sends it to the server via their terminal, the server receives the message and saves it as a log. The message is then analyzed by a natural language processing engine, and tokenization and keyword extraction are performed. The extracted keywords are classified using a machine learning model, and appropriate countermeasures are searched for in a database based on the obtained alarm type. At the same time, the emotion engine analyzes the user's emotions. The way the countermeasures are presented is adjusted based on the emotional information. Finally, countermeasure information and an encouraging message corresponding to the user's emotions are displayed to the user via their terminal.

[1165] This system solves the problem of difficulty in understanding alarm messages and enables appropriate and prompt responses according to the user's emotional state. For example, if the message "Power supply failure detected at base station 5" is input, the system will classify this as a "power supply failure" and provide measures such as "check the backup power supply and replace it if necessary," as well as an encouraging message such as "please stay calm and continue working. The problem can be solved." As a result, the user can carry out their work efficiently and with less mental strain.

[1166] The processing flow will be explained below.

[1167] Step 1:

[1168] The user inputs an alarm message into the terminal, for example, "Power supply failure detected at base station 5."

[1169] Step 2:

[1170] The terminal sends the entered alarm message to the server. The message is sent to the server as a character string.

[1171] Step 3:

[1172] The server receives alarm messages sent from the terminals and records the received messages in a database as a log.

[1173] Step 4:

[1174] The server passes the received message to a natural language processing engine (NLP engine), which tokenizes the message and extracts keywords and important phrases, such as "Power supply failure" and "base station 5."

[1175] Step 5:

[1176] The server inputs the extracted keywords into a machine learning model, which then classifies the alarm type based on these keywords. For example, the alarm may be classified as "power supply failure" based on the keywords.

[1177] Step 6:

[1178] Based on the alarm classification, the server searches for the appropriate countermeasure from a database that records past response cases and corrective actions.

[1179] Step 7:

[1180] To recognize the emotions expressed when a user inputs an alarm message, the server runs an emotion engine. The emotion engine extracts emotion data from the input message and user operations, and analyzes the user's emotional state, such as "anxiety" or "tension."

[1181] Step 8:

[1182] The server analyzes the emotional information recognized by the emotion engine and adjusts the way the countermeasures are presented. If the user is feeling strong stress, it will use calmer and more positive expressions.

[1183] Step 9:

[1184] The server converts the detected countermeasure information and adjusted information based on it into a format for display in the user interface. For example, in addition to the countermeasure "Check the backup power supply and replace it if necessary," it generates an encouraging message such as "Please stay calm and continue working. The problem can be solved."

[1185] Step 10:

[1186] The server sends the generated formatted countermeasure information and the adjusted information to the terminal.

[1187] Step 11:

[1188] The device receives countermeasure information from the server and displays advice based on the user's emotion. The user can see information on the screen such as "Alarm type: Power supply failure," "Corrective action: Check the backup power supply and replace it if necessary," and "Please remain calm and continue working. The problem can be solved."

[1189] Step 12:

[1190] The user can take appropriate action based on the displayed instructions and encouraging messages, thereby quickly implementing measures to resolve the cause of the alarm.

[1191] Example 2

[1192] Next, a description will be given of 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."

[1193] Conventional alarm message analysis systems analyzed alarm messages and proposed appropriate countermeasures, but because they ignored the emotions felt by users when inputting messages, the countermeasures were sometimes not accepted or implemented smoothly. Furthermore, the lack of appropriate support for emotions such as anxiety and tension felt by users led to problems such as a heavy mental burden on users and reduced work efficiency. To solve these problems, a system that not only analyzes alarm messages but also takes into account the user's emotions is needed.

[1194] 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.

[1195] In this invention, the server includes means for saving alarm messages as logs, means having an emotion analysis engine for analyzing emotions from text entered by a user, and means for adjusting the manner in which countermeasures are presented based on the analyzed emotion information. This not only analyzes the contents of alarm messages and proposes appropriate countermeasures, but also adjusts the manner in which the countermeasures are presented according to the emotions felt by the user, thereby reducing mental burden and improving work efficiency.

[1196] An "alarm message" is a message used to report problems or abnormalities that occur on the system.

[1197] "User interface" refers to the screens and input devices that allow a user to interact with a system.

[1198] A "server" is a computer system that processes and stores data over a network.

[1199] A "natural language processing engine" is a technology that analyzes human language and allows computers to understand and analyze it.

[1200] "Tokenization" is the process of dividing a sentence into units such as words and phrases.

[1201] "Keywords" are words or phrases that are particularly important in a piece of text.

[1202] A "machine learning model" is an algorithm that learns from data and makes predictions and classifications.

[1203] A "database" is a system for efficiently managing and searching large amounts of data.

[1204] An "emotion analysis engine" is a technology that analyzes and identifies emotions from text data.

[1205] A "log" is recorded data that a system generates during operation.

[1206] The present invention is a system for analyzing alarm messages related to base station construction and proposing countermeasures, and is configured to propose countermeasures taking into consideration the user's emotions. This system is composed of a user interface, a natural language processing engine installed on a server, a machine learning model, a sentiment analysis engine, and a database.

[1207] First, the user inputs an alarm message through the terminal. For example, if the user inputs the message "Power supply failure detected at base station 5," the terminal sends this message to the server. The server receives the message and stores it in a database as a log. The stored message is then analyzed by a natural language processing engine and tokenized into words and phrases. In this case, keywords such as "Power supply failure" and "base station 5" are extracted.

[1208] The extracted keywords are then fed into a machine learning model, which uses a pre-trained algorithm to classify the alarm type based on the keywords. For example, it may be classified as "power supply failure." The server then searches a database for an appropriate countermeasure for the classified alarm. For example, it may find a countermeasure such as "check the backup power supply and replace it if necessary."

[1209] Furthermore, an emotion analysis engine analyzes the emotions from the text entered by the user. The emotion analysis engine identifies emotions such as "anxiety" or "tension" felt by the user. Based on this emotional information, the way the countermeasures are presented is adjusted. For example, if the user is feeling very anxious, the server will add encouraging words such as "Please stay calm and continue working. The problem can be solved."

[1210] Finally, the adjusted countermeasures and advice are displayed to the user via the terminal. As a specific example, if the user inputs the alarm message "Power supply failure detected at base station 5," the system will classify this as a "power supply failure" and display the countermeasure "Check the backup power supply and replace it if necessary," along with encouraging words such as "Please stay calm and work. The problem can be solved." An example of this prompt would be "Enter the alarm message and output the analysis results and countermeasures."

[1211] This system not only improves analysis accuracy compared to conventional systems, but also takes the user's emotions into consideration, reducing mental burden and enabling more appropriate countermeasures to be provided more quickly.

[1212] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1213] Step 1:

[1214] The user enters an alarm message

[1215] The user inputs an alarm message into the input field of the terminal. For example, the user inputs "Power supply failure detected at base station 5." This input is sent to the terminal as text data.

[1216] Step 2:

[1217] The device sends a message to the server

[1218] The device sends the input message to the server using an HTTP request, which is passed to the server as text data. At this time, the data is encoded into a common data format such as JSON.

[1219] Step 3:

[1220] The server receives the message and stores it in a log.

[1221] The server receives the HTTP request and stores it in temporary memory. Then, it saves the message as a log in a database. A database management system such as MySQL or PostgreSQL is used for saving. The input data is in text format, and the output is saved as a log file or a record in the database.

[1222] Step 4:

[1223] The server analyzes the message using a natural language processing engine

[1224] The server inputs the stored messages into a natural language processing engine. For example, it uses Python's NLTK library to tokenize the messages, which extracts keywords such as "Power supply failure" and "base station 5." The input is text data, and the output is a list of keywords.

[1225] Step 5:

[1226] The server classifies the alarm type using a machine learning model.

[1227] The server inputs the extracted keywords into a pre-trained machine learning model, for example, Scikit-learn, to classify the alarm type. The input is a list of keywords, and the output is an alarm type, such as "power supply failure."

[1228] Step 6:

[1229] The server searches the database for the appropriate solution

[1230] The server searches the database for appropriate countermeasures based on the classified alarm type. The database stores past alarm response history and a knowledge base. The input is the alarm type, and the output is a countermeasure such as "check the standby power supply and replace it if necessary."

[1231] Step 7:

[1232] The server analyzes the user's emotions using an emotion analysis engine.

[1233] The server uses a sentiment analysis engine to analyze the sentiment of the user's input text. For example, it uses Python's TextBlob library to identify emotions such as "anxiety" or "tension" that the user is feeling. The input is text data, and the output is an emotion label.

[1234] Step 8:

[1235] The server adjusts how the measures are expressed.

[1236] The server adjusts the presentation of the countermeasures it provides based on the analyzed emotion information. For example, if the user feels "anxious," it adds encouraging words such as "Please stay calm and work. The problem can be solved." The input is the emotion label and countermeasure data, and the output is the adjusted countermeasure message.

[1237] Step 9:

[1238] The device displays countermeasures and advice to the user

[1239] The terminal displays the countermeasure information sent from the server and the adjusted advice to the user. For example, the terminal displays "Check the backup power supply and replace it if necessary. Please stay calm and work. The problem can be solved." The input is the adjusted countermeasure message, and the output is the display on the user interface.

[1240] (Application example 2)

[1241] Next, a description will be given of Application 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."

[1242] In conventional factory robot operations, it has been difficult to immediately analyze error messages from the robot and provide appropriate countermeasures. Furthermore, if a robot operator feels anxious or stressed by an error message, there is a lack of means to provide support that responds to their emotions. This makes it difficult for operators to work efficiently and with peace of mind, and there is a risk of equipment stalling or incorrect operation.

[1243] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means having a user interface for inputting an alarm message; means equipped with a natural language processing engine that receives the alarm message, tokenizes it, and extracts keywords; means including a machine learning model that classifies the alarm type based on the keywords; means for searching a database for appropriate countermeasures for the alarm type and displaying the results on the user interface; means including an emotion analysis engine that recognizes the emotion of the user when entering the message; and means for adjusting the way advice is presented based on the results of the emotion analysis engine. This makes it possible to quickly and accurately analyze error messages from factory robots and provide appropriate countermeasures, as well as provide appropriate support according to the operator's emotional state.

[1244] Below are definitions of key terms found in the patent claims:

[1245] An "alarm message" refers to a notification sent when a device such as a robot or machine detects an abnormality or error.

[1246] "User interface" refers to the means or screen through which a user directly interacts with a computer system or application.

[1247] A "natural language processing engine" refers to the technology or system that enables a computer to understand and analyze human language and extract meaning.

[1248] "Tokenization" refers to the process of dividing a sentence into words and phrases.

[1249] A "keyword" is a word that has a particularly important meaning in a sentence or paragraph.

[1250] A "machine learning model" refers to an algorithm or system that learns patterns from data and uses those patterns to make inferences and classify new data.

[1251] A "database" refers to a system that systematically stores and manages data, allowing it to be quickly searched and used as needed.

[1252] An "emotion analysis engine" refers to a system that recognizes and analyzes emotions from user input, tone of voice, facial expressions, etc.

[1253] "Adjusting advice presentation" refers to changing the wording or presentation of advice or instructions provided depending on the user's emotional state.

[1254] This invention is a system that allows factory robot operators to analyze error messages from the robot and quickly provide appropriate countermeasures. Furthermore, it can improve work efficiency by recognizing the operator's emotions and providing appropriate support according to those emotions. This system includes the following components:

[1255] 1. Overall system configuration

[1256] The overall system configuration is as follows:

[1257] User interface: A means for operators to input error messages. This can be a factory robot control terminal, a smartphone, or a tablet device.

[1258] Server: A computer system that performs the main processing for analyzing error messages and providing appropriate countermeasures. The server is equipped with a natural language processing engine, machine learning models, a sentiment analysis engine, and a database.

[1259] Natural Language Processing Engine: A library for tokenizing error messages and extracting keywords, for example using Python's TextBlob library.

[1260] Machine learning model: A model for classifying error message types based on extracted keywords. This model is built using the sklearn library.

[1261] Sentiment analysis engine: An engine for recognizing emotions from operator input. For sentiment analysis, the transformers library is used, along with the BERT model.

[1262] Database: A source of information that accumulates solutions to error messages. This database stores past error messages and their corresponding solutions.

[1263] 2. System Operation

[1264] The flow for this system to work properly is as follows:

[1265] Alarm message entry and analysis

[1266] When an operator enters an error message on the factory robot's control terminal, the message is sent to the server. The server receives the message and saves it as a log. The natural language processing engine then analyzes the message and extracts the necessary keywords. For example, if the message entered is "Power supply failure detected at manufacturing robot 3," the words "Power supply failure" and "manufacturing robot 3" are extracted.

[1267] Error message classification

[1268] The extracted keywords are input into a machine learning model to classify the type of error message. For example, this one might be classified as "power supply failure." For each classified error message, the server accesses a database to search for the appropriate solution.

[1269] Sentiment analysis and advice provision

[1270] The emotion analysis engine analyzes the operator's emotions, identifying emotions such as "anxiety" or "stress" from input such as "This is really stressing me out." Based on this analysis result, the way countermeasures are presented is adjusted. For example, if the operator is feeling anxious, a positive message such as "Please stay calm and continue working. The problem can be solved" is added.

[1271] 3. Specific Examples

[1272] As a specific example, if an operator inputs the alarm message "Power supply failure detected at manufacturing robot 3" and the emotion input "This is really stressing me out," the system will classify this as a "power supply failure" and provide a solution: "Check the backup power supply and replace it if necessary." At the same time, it will also display an encouraging message: "Please stay calm and continue working. The problem can be solved."

[1273] Prompt Sentence Examples

[1274] "I'm experiencing a power supply failure. What's the best way to resolve this issue?"

[1275] "Please advise how to respond if users are feeling uneasy."

[1276] This system enables factory robot operators to deal with error messages quickly and appropriately, while at the same time reducing the psychological burden.

[1277] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1278] Step 1:

[1279] The user enters an error message.

[1280] Specific operation: The user uses the factory robot's management terminal or smartphone to enter an error message such as "Power supply failure detected at manufacturing robot 3."

[1281] Input: Error message

[1282] Output: The entered error message

[1283] Step 2:

[1284] The terminal sends an error message to the server.

[1285] Specific operation: The error message entered by the user is automatically sent from the management terminal to the server.

[1286] Input: Error message

[1287] Output: Error message sent to the server

[1288] Step 3:

[1289] The server receives the error message and stores it in a log.

[1290] Specific behavior: The server records the received error message in a database for later retrieval.

[1291] Input: The error message sent

[1292] Output: Error messages stored in the database

[1293] Step 4:

[1294] The server uses a natural language processing engine to tokenize the error message and extract keywords.

[1295] What it does: It uses Python's TextBlob library to analyze error messages and extract keywords such as "Power supply failure" and "manufacturing robot 3."

[1296] Input: Saved error message

[1297] Output: Extracted keywords

[1298] Step 5:

[1299] The server uses a machine learning model to classify the type of error message based on keywords.

[1300] Specific operation: The extracted keywords are input into a model built with the sklearn library and classified into error message types such as "power supply failure."

[1301] Input: Extracted keywords

[1302] Output: Classified error message types

[1303] Step 6:

[1304] The server searches the database for an appropriate solution based on the type of classified error message.

[1305] Specific operation: For an error message classified as "power supply failure," the database is searched for countermeasures such as "check the backup power supply and replace it if necessary."

[1306] Input: Classified error message type

[1307] Output: Measures found

[1308] Step 7:

[1309] The user inputs emotions.

[1310] Specific operation: The user inputs a feeling such as "This is really stressing me out" into the management terminal.

[1311] Input: Emotion input

[1312] Output: Input emotion text

[1313] Step 8:

[1314] The server analyzes the user's emotions using an emotion analysis engine.

[1315] Specific behavior: Analyzes input emotion text using the BERT model from the transformers library to identify emotions such as "anxiety" and "stress."

[1316] Input: Entered emotion text

[1317] Output: Parsed sentiment

[1318] Step 9:

[1319] The server adjusts the way advice is presented based on the emotions analyzed.

[1320] Specific action: If the analyzed emotion is "anxiety," add an encouraging message to the searched measures: "Please stay calm and work. The problem can be solved."

[1321] Input: Parsed sentiment, retrieved measures

[1322] Output: Adjusted advice

[1323] Step 10:

[1324] The server displays the tailored advice in a user interface.

[1325] Specific operation: Display the adjusted advice on the management terminal screen and provide the user with countermeasures and encouraging messages.

[1326] Input: Tailored Advice

[1327] Output: Advice displayed

[1328] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding 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.

[1329] 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> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with 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.

[1330] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1331] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1332] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1333] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1334] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1335] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1336] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1337] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1338] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1339] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1340] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1341] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1342] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1343] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1344] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1345] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1346] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1347] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1348] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1349] The following is further disclosed regarding the above embodiment.

[1350] (Claim 1)

[1351] means having a user interface for inputting an alarm message;

[1352] means for receiving the alarm message, and tokenizing and extracting keywords from the alarm message, the means comprising a natural language processing engine;

[1353] means including a machine learning model for classifying alarm types based on the keywords;

[1354] means for searching a database for an appropriate measure for the type of alarm and displaying the result on a user interface;

[1355] A system including:

[1356] (Claim 2)

[1357] 10. The system of claim 1, further comprising a log storage means for recording data on receiving an alarm message.

[1358] (Claim 3)

[1359] 2. The system according to claim 1, further comprising means for analyzing the cause of an alarm based on keywords extracted by the natural language processing engine.

[1360] "Example 1"

[1361] (Claim 1)

[1362] means for a user to input an alarm message through a terminal;

[1363] means for receiving the alarm message and storing it as a log;

[1364] means for tokenizing the alarm message using a natural language processing engine and extracting keywords;

[1365] A means for inputting the extracted keywords into a machine learning model to classify the alarm type;

[1366] means for searching for appropriate countermeasures from a database based on the classified alarms and displaying the results to a user through a user interface;

[1367] A system including:

[1368] (Claim 2)

[1369] 10. The system of claim 1, further comprising a log storage means for recording data on receiving an alarm message.

[1370] (Claim 3)

[1371] 2. The system according to claim 1, further comprising means for analyzing the cause of an alarm based on keywords extracted by the natural language processing engine.

[1372] "Application Example 1"

[1373] (Claim 1)

[1374] means having a user interface for inputting an alarm message;

[1375] means for receiving the alarm message, and tokenizing and extracting keywords from the alarm message, the means comprising a natural language processing engine;

[1376] means including a machine learning model for classifying alarm types based on the keywords;

[1377] means for searching a database for an appropriate measure for the type of alarm and displaying the result on a user interface;

[1378] A means for transmitting an alarm message from a device equipped with an abnormality detection sensor in a factory;

[1379] The system further comprises means for analyzing the alarm message and suggesting appropriate countermeasures to a factory worker.

[1380] (Claim 2)

[1381] 10. The system of claim 1, further comprising a log storage means for recording data on receiving an alarm message.

[1382] (Claim 3)

[1383] 2. The system according to claim 1, further comprising means for analyzing the cause of an alarm based on keywords extracted by the natural language processing engine.

[1384] "Example 2: Combining Emotion Engines"

[1385] (Claim 1)

[1386] means having a user interface for inputting an alarm message;

[1387] means for receiving the alarm message, and tokenizing and extracting keywords from the alarm message, the means comprising a natural language processing engine;

[1388] means including a machine learning model for classifying alarm types based on the keywords;

[1389] means for searching a database for an appropriate measure for the type of alarm and displaying the result on a user interface;

[1390] means for storing the alarm message as a log;

[1391] means for analyzing emotions from a user's input text; and

[1392] a means for adjusting a method of expressing measures to be provided based on the analyzed emotion information;

[1393] A system including:

[1394] (Claim 2)

[1395] 10. The system of claim 1, further comprising a log storage means for recording data on receiving an alarm message.

[1396] (Claim 3)

[1397] 2. The system according to claim 1, further comprising means for analyzing the cause of an alarm based on keywords extracted by the natural language processing engine.

[1398] "Application example 2 when combining emotion engines"

[1399] (Claim 1)

[1400] means having a user interface for inputting an alarm message;

[1401] means for receiving the alarm message, and tokenizing and extracting keywords from the alarm message, the means comprising a natural language processing engine;

[1402] means including a machine learning model for classifying alarm types based on the keywords;

[1403] means for searching a database for an appropriate measure for the type of alarm and displaying the result on a user interface;

[1404] means including an emotion analysis engine for recognizing the emotion of a user's input;

[1405] means for adjusting the way advice is expressed based on the result of the sentiment analysis engine;

[1406] A system including:

[1407] (Claim 2)

[1408] 10. The system of claim 1, further comprising a log storage means for recording data on receiving an alarm message.

[1409] (Claim 3)

[1410] 2. The system according to claim 1, further comprising means for analyzing the cause of an alarm based on keywords extracted by the natural language processing engine. [Explanation of symbols]

[1411] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means having a user interface for inputting an alarm message; means for receiving the alarm message, and tokenizing and extracting keywords from the alarm message, the means comprising a natural language processing engine; means including a machine learning model for classifying alarm types based on the keywords; means for searching a database for an appropriate measure for the type of alarm and displaying the result on a user interface; A system including:

2. 2. The system of claim 1, further comprising a log storage means for recording data on received alarm messages.

3. The system according to claim 1 , further comprising means for analyzing the cause of an alarm based on keywords extracted by the natural language processing engine.

Images