System

Abstract

A system is provided.SOLUTION: The system includes a means for acquiring examination number data from a past database, a means for acquiring weather information from an external weather forecast service, a means for acquiring event information from an external event information service, a means for preprocessing the acquired data, a means for training a prediction model using the preprocessed data, and a means for predicting future examination numbers.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including 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] Conventional systems for predicting the number of screening cases make predictions based solely on past data, and are unable to take into account external factors such as weather or events. This makes it difficult to accurately predict fluctuations in screening cases due to seasons or specific events. Furthermore, manually taking these factors into account is extremely time-consuming and labor-intensive. This can lead to problems such as reduced prediction accuracy and misallocation of resources. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system for predicting the number of inspections based not only on past inspection number data but also on weather forecasts and various event information. Specifically, the system includes a means for acquiring inspection number data from a past database, a means for acquiring weather information from an external weather forecast service, a means for acquiring event information from an external event information service, a means for preprocessing the acquired data, a means for training a prediction model using the preprocessed data, and a means for predicting the number of future inspections. This system enables highly accurate predictions of the number of inspections that take external factors such as weather and events into account, thereby achieving optimal resource allocation.

[0006] A "means" is a method, apparatus, program, or process for performing a particular function or action.

[0007] A "database" is a collection of information for efficiently storing, managing, and retrieving data, and is often managed by a relational database management system.

[0008] "Inspection number data" is information indicating the number and volume of inspections conducted during a specific period.

[0009] A "weather forecast service" is a service that predicts and provides future weather conditions.

[0010] "Weather information" refers to data and forecast information related to weather, such as temperature, humidity, precipitation, and wind speed.

[0011] An "event information service" is a service that provides information on events taking place in a specific region or period.

[0012] "Event information" is data about an event, such as the event name, date, location, and number of participants.

[0013] "Preprocessing" refers to a series of preparatory steps to prepare data for analysis and model training.

[0014] A "predictive model" is an algorithm or mathematical model for predicting future values ​​based on past data.

[0015] A "linear regression model" is a statistical method that learns the linear relationship between input data and output data and makes predictions based on that.

[0016] A "system" is a collection of interrelated elements that perform a specific function, including hardware, software, and communication infrastructure. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] This system consists of a server and a user terminal. The server acquires the necessary data from weather forecast services and event information services and performs a series of processes to train the model. The user terminal provides an interface for users to check the results.

[0039] Explanation of program processing

[0040] Data collection

[0041] 1. The server first retrieves audit data from the historical audit database, which involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame.

[0042] 2. The server sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. The weather information includes data such as temperature, humidity, and weather type for each date.

[0043] 3. The server also sends an HTTP request to an external event information service to retrieve event information for the specified region and period, including the event name, date, location, and number of participants.

[0044] Data Preprocessing

[0045] 4. The server combines the acquired inspection data, weather information, and event information to generate a unified dataset. Each data frame is merged based on date.

[0046] 5. The server will gracefully handle missing values ​​and outliers in the combined datasets, using imputation such as zeros and mean values.

[0047] Training the model

[0048] 6. The server uses the preprocessed data to train a linear regression model, which uses weather and event information as features and the number of reviews as targets.

[0049] 7. The server stores the trained model for future predictions.

[0050] Prediction of future review volume

[0051] 8. The server retrieves future weather and event information, preprocesses it, and reassembles it into a dataset ready to be fed into the model.

[0052] 9. The server uses the trained linear regression model to predict future audit volume. The prediction results are added to a data frame and can be saved or sent to a device.

[0053] Specific examples

[0054] For example, to predict next month's number of inspections, the server first retrieves data on the number of inspections over the past year. Then, it collects weather and event information for the same period along the time axis, integrates this data, and generates a training dataset. It then trains a linear regression model to predict the number of inspections based on next month's weather and event information. Finally, users can view the results on their devices.

[0055] Users can use the prediction results to, for example, efficiently allocate staff shift schedules or adjust response resources in advance. In this way, the system improves the accuracy of predictions of the number of examination cases, contributing to operational efficiency.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] The server connects to the database of past review counts. It uses a database connection library, such as sqlite3 or sqlalchemy, to connect to the database. It issues an SQL query to retrieve past review count data from the database. Specifically, it executes a query such as SELECT FROM review_counts and stores the results in a Pandas data frame.

[0059] Step 2:

[0060] The server sends an HTTP request to the weather forecast service. Using the requests library, it sends parameters including the API key, location information, and forecast period (number of days) to the weather forecast service's API endpoint (e.g., https: / / api.weatherapi.com / v1 / forecast.json). It then obtains the JSON data returned as a response, extracts the necessary weather information (date, time, temperature, weather type, etc.), and converts it into a data frame.

[0061] Step 3:

[0062] The server also sends an HTTP request to the event information service. Using the requests library, it sends parameters including the API key, location information, and duration to the event information service's API endpoint (e.g., https: / / api.eventapi.com / v1 / events). The necessary information (event name, date and time, location, number of attendees, etc.) is extracted from the JSON data returned as a response and converted into a data frame.

[0063] Step 4:

[0064] The server combines the obtained inspection count data, weather information, and event information. It uses the Pandas merge function to combine each data frame using the date as a key. This operation generates a single combined dataset.

[0065] Step 5:

[0066] The server processes missing values ​​in the integrated dataset. It uses Pandas' fillna function to fill missing values ​​with zeros or the average of the preceding and following days. This process ensures the completeness of the dataset.

[0067] Step 6:

[0068] The server trains a linear regression model using the preprocessed data. Using Scikit-learn's LinearRegression class, it separates the features (weather information, event information) and the target (number of reviews) and trains using the model's fit method. Through this training, the model learns parameters from the given data.

[0069] Step 7:

[0070] The server saves the trained model for future predictions. The model is saved using the joblib or pickle libraries, and stored as a file on the local disk or in a database.

[0071] Step 8:

[0072] The server again obtains the weather and event information needed for future predictions. It also sends an HTTP request to obtain future data and performs preprocessing. The obtained data is again incorporated into the integrated dataset using the date as a key.

[0073] Step 9:

[0074] The server uses the trained model to predict future audit volume. It inputs the preprocessed future data into the model and gets the prediction result. The result is added as a new column in the data frame.

[0075] Step 10:

[0076] The server saves the prediction results in a database or sends them to the user's device. When saving the results in a database, an INSERT operation is performed using an appropriate SQL query. When notifying the user, the server sends the results via email or a push notification service, for example.

[0077] Step 11:

[0078] Users can check the forecast results on their devices, view the predicted number of reviews via a dedicated management screen or notification messages, and adjust staff schedules and create work plans based on the results.

[0079] Example 1

[0080] 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."

[0081] Conventional systems had low accuracy in predicting the number of screening cases, making it difficult to improve work efficiency. Furthermore, a series of processes, such as collecting data from external services, preprocessing the data, and training the prediction model, were often performed manually, resulting in a lot of work and prone to errors. Therefore, a more accurate and automated prediction system was needed.

[0082] 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.

[0083] In this invention, the server includes a means for acquiring case count data from a past database, a means for acquiring weather data from an external information service, and a means for acquiring event information from an external information service. This enables the acquired data to be automatically integrated and preprocessed. Furthermore, the preprocessed data can be used to train a prediction model, and future case counts can be predicted with high accuracy based on the prediction results. This improves business efficiency and accuracy.

[0084] The "past database" is a data storage that holds the number of past reviews and related data.

[0085] "Case data" is data that includes information regarding the number of reviews conducted within a particular period of time.

[0086] An "external information service" is an external service that the server accesses via the Internet and that provides weather-related data and forecasts.

[0087] "Weather data" is data that includes information about the weather in a particular region or time period, such as temperature, humidity, and type of weather.

[0088] An "external information service" is an external service that the server accesses via the Internet and provides information about events and other activities.

[0089] "Event information" is data that includes details related to an event, such as the event name, date, location, and number of participants.

[0090] "Preprocessing means" refers to a set of operations or techniques used to process acquired data by cleansing, filtering, integrating, etc., and converting it into a form suitable for training a predictive model.

[0091] "Means of integration" refers to a method of combining data obtained from different sources using dates as a key to create a single unified data set.

[0092] A "predictive model" is a mathematical or statistical model trained to predict future values ​​based on past data.

[0093] A "regression analysis model" is a statistical method for analyzing the relationship between independent variables and dependent variables and making predictions.

[0094] "Future volume" refers to the number of audits that will occur under specified conditions in a projected future period.

[0095] A "server" is a computer system that executes a series of processes such as data acquisition, preprocessing, model training, and prediction calculation.

[0096] The present invention is a system that obtains event count data from a past database, weather data and event information from external information services, and predicts future event counts based on this data. The system is composed of a server and a user terminal.

[0097] Server configuration and functions

[0098] 1. Data Acquisition

[0099] The server first retrieves the historical data from the database. This is done using a database connection library and an SQL query. In Python, the pandas read_sql_query function is used to retrieve the data in data frame format.

[0100] Next, the server retrieves weather data from an external information service (e.g., a weather forecast service API) by sending an HTTP request using the requests library and parsing the resulting data in JSON format.

[0101] Additionally, the server obtains event information from an external event information service (e.g., an event information API). Again, it uses the requests library to send an HTTP request and parses the returned data in JSON format.

[0102] 2. Data Preprocessing

[0103] The server converts the acquired count data, weather data, and event information into respective data frames and combines them based on the date. This combining process uses the pandas merge function. After combining, the data set is filled in with missing values ​​and outliers are processed. For the filling process, the pandas fillna method is used to fill in missing values ​​with zero or the average value.

[0104] 3. Training the model

[0105] The server trains a predictive model (regression analysis model) based on the preprocessed dataset. For training, it uses the LinearRegression class from scikit-learn, learning the model using weather data and event information as features and count data as the target. The trained model is then serialized and saved using the joblib library.

[0106] 4. Future Data Prediction

[0107] The server again retrieves and preprocesses future weather and event data. Based on this new data, it uses the trained regression analysis model to predict future event counts. The prediction results are sent to the user's device or stored locally on the server.

[0108] User terminal configuration and functions

[0109] The user terminal provides an interface for displaying the forecast results sent from the server, allowing the user to improve work efficiency and optimize resource allocation based on these forecast figures.

[0110] Specific examples

[0111] For example, when making a prediction for next month, the server retrieves the number data for the past year from the database. At the same time, it retrieves weather data for the specified period from the weather forecast service API and collects event information from the event information provision API. This data is integrated, missing values ​​are filled in, and a training dataset is generated. The server then trains a regression analysis model based on this dataset and predicts the number of events for next month by inputting future weather data and event information. Finally, users can check these prediction results on their own devices and use them in their work.

[0112] Prompt Sentence Examples

[0113] "Please predict the number of cases next month based on the number of cases over the past year, as well as weather and event information for the specified area."

[0114] As described above, the present invention provides a system that makes highly accurate predictions based on past data and external information, thereby improving business efficiency.

[0115] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0116] Step 1:

[0117] The server first connects to the historical database and retrieves the count data. Specifically, it issues an SQL query using a database connection library. For example, it executes the query "SELECT FROM Review Count WHERE Date BETWEEN (registered trademark) '2022-01-01' AND '2022-12-31'" and retrieves the results in data frame format. The input is an SQL query, and the output is count data in data frame format.

[0118] Step 2:

[0119] The server obtains weather data from an external information service. Specifically, it uses the requests library to send an HTTP request to obtain weather information for a specified region and period. The request URL includes parameters such as the API key, region, and period. For example, it is executed as follows: "requests.get('https: / / api.weather.com / v3 / wx / forecast / daily / 5day?apiKey=YOUR_API_KEY&geocode=LAT,LON')". The input is the API request, and the output is weather data in JSON format.

[0120] Step 3:

[0121] The server obtains event information from an external information provider. Again, it uses the requests library to send an HTTP request to obtain event information for a specified region and period. For example, execute it as follows: "requests.get('https: / / api.eventservice.com / v1 / events?apiKey=YOUR_API_KEY&location=LOCATION&startDate=START_DATE&endDate=END_DATE')". The input is the API request, and the output is event information in JSON format.

[0122] Step 4:

[0123] The server integrates the acquired count data, weather data, and event information. To do this, it uses the pandas merge function to combine each data frame based on the date. For example, execute "merged_data = pd.merge(count data, weather data, on='date')". The input is the count data, weather data, and event information in data frame format, and the output is an integrated dataset.

[0124] Step 5:

[0125] The server processes missing values ​​and outliers in the merged dataset. Specifically, it uses the pandas fillna method to fill missing values. Zero or the mean value is used for this filling. For example, execute "merged_data.fillna(0, inplace=True)". The input is the merged dataset, and the output is the imputed dataset.

[0126] Step 6:

[0127] The server trains a regression analysis model using the preprocessed dataset. Using scikit-learn's LinearRegression class, the model is trained using weather data and event information as features and count data as the target. For example, this is done using "from sklearn.linear_model import LinearRegression", "model = LinearRegression()", and "model.fit(features, target)". The input is the preprocessed dataset, and the output is a trained regression analysis model.

[0128] Step 7:

[0129] The server saves the trained model. It serializes and saves the model using the joblib library. For example, run "import joblib" and "joblib.dump(model, 'model.pkl')". The input is the trained regression model, and the output is the saved model file.

[0130] Step 8:

[0131] The server retrieves future weather data and event information again and performs preprocessing. It converts them into a data frame format again and merges them based on the date. For example, execute "future_data = pd.merge(future weather data, future event data, on='date')". The input is future weather data and event information, and the output is a preprocessed future dataset.

[0132] Step 9:

[0133] The server uses the trained regression analysis model to predict future count data. Future weather data and event information are input as features into the model to obtain prediction results. For example, execute "future_predictions = model.predict(future features)". The input is a preprocessed future dataset, and the output is the predicted future count data.

[0134] Step 10:

[0135] The server adds the prediction results to a data frame and sends it to the user's device or saves it in an appropriate database. For example, it can save it as a CSV file using the pandas to_csv method or send it to the user's device through an API. The input is a data frame containing the prediction results, and the output is the saved file or the sent data.

[0136] The above are the processing steps of the program for this system.

[0137] (Application example 1)

[0138] 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."

[0139] In traditional brick-and-mortar store operations, it was difficult to accurately predict the number of customers because operations depended on weather and local event information, making it difficult to optimize staff allocation and inventory management. This also led to staff shortages and overstaffing, hindering efficient store operations. This resulted in increased costs and a decline in service quality.

[0140] 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.

[0141] In this invention, the server includes means for acquiring data from a past database, means for acquiring information from an external weather information service, means for acquiring activity information from an external activity information service, means for preprocessing the acquired data, means for training a prediction model using the preprocessed data, means for predicting the number of customers in the future, means for optimizing staffing using the predicted number of customers in the future, and user interface means for displaying the optimized staffing and inventory management. This makes it possible to accurately predict the number of customers in the future and optimally allocate staff and inventory.

[0142] A "database" is a system designed to efficiently organize and manage information and data.

[0143] "Data acquisition means" refers to the methods and processes used to collect the necessary data.

[0144] "Weather information service" means an external source that provides information about the weather and climate for a particular region or period of time.

[0145] "Activity information services" are external information sources that provide information on local events and activities.

[0146] "Preprocessing" refers to the process of converting and organizing data into a format suitable for analysis and model training.

[0147] A "predictive model" is a computational algorithm for predicting future situations based on past data and factors.

[0148] "Customer traffic forecasting method" means a method or process used to forecast future customer traffic.

[0149] "Staffing optimization measures" are methods and techniques for most efficient staffing.

[0150] A "user interface" is a part of a system that provides a means of input and output that allows a user to directly interact with the system.

[0151] "Store foot traffic" refers to the number of customers who visit a physical store within a specific period of time.

[0152] To implement this invention, a system consisting of a server and a user terminal must be constructed. The server acquires data from a past database and collects information from external weather information services and external activity information services. The user terminal acts as an interface that provides users with forecast results, optimized staffing, and inventory management information.

[0153] ---

[0154] 1. Data Collection

[0155] The server first retrieves customer visit data from a historical database by issuing SQL queries using a database connectivity library and retrieving the results as a data frame, and then retrieves weather and event information using HTTP requests from external weather information services (e.g., OpenWeatherMap API) and activity information services (e.g., Eventful API).

[0156] ---

[0157] 2. Data Preprocessing

[0158] The acquired customer visit data, weather information, and event information are integrated based on the date, and missing or outlier values ​​in the integrated dataset are filled with zeros or the average value.

[0159] ---

[0160] 3. Model training

[0161] A linear regression model is trained using the preprocessed data. This process uses Python's Scikit-learn library. Using weather and event information as features and the number of customer visits as targets, a model is constructed to predict future customer visits.

[0162] ---

[0163] 4. Future Predictions

[0164] Future weather and event information is taken, combined into a preprocessed dataset, and a trained linear regression model is used to predict future store traffic. The prediction results are added to a data frame and saved or sent to the device as a result.

[0165] ---

[0166] 5. User Interface

[0167] The user device displays predicted future customer numbers, optimized staffing, and inventory management information. Staffing plans and inventory management can be visually confirmed using a smartphone app or the robot's display.

[0168] ---

[0169] Specific examples

[0170] For example, to generate a forecast for the next month, the server first retrieves customer visitation data from the past year. Next, it collects weather and event information from the same period along the time axis, integrates this data, and generates a training dataset. It then trains a linear regression model to predict the number of customers based on the weather and event information for the next month. Finally, users can view the results on their devices and, for example, efficiently allocate staff shift schedules or adjust response resources in advance.

[0171] ---

[0172] Prompt Sentence Examples

[0173] "To predict the number of customers coming to Store A next month, we will provide the following information: number of customers visiting the store over the past year, weather information (temperature, humidity, weather type), and event information (event name, date, location, number of participants). Based on this information, please calculate the predicted number of customers coming to the store next month."

[0174] This system is expected to significantly improve the efficiency of physical store operations, as it not only predicts store visits but also enables efficient staff deployment planning and optimized inventory management.

[0175] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0176] Step 1:

[0177] The server retrieves historical customer visit data from a database. Specifically, it uses a database connection library to issue an SQL query and obtains the results as a data frame. This data includes the number of customer visits by date. The input is the database connection information and the SQL query, and the output is a data frame of the number of customer visits by date.

[0178] Step 2:

[0179] The server obtains weather information from an external weather information service. Specifically, it sends an HTTP request to the weather information API and receives weather information as a response. This weather information includes temperature, humidity, and weather type by date. The input is the API key and the request URL, and the output is a data frame of weather information by date.

[0180] Step 3:

[0181] The server obtains event information from an external activity information service. Specifically, it sends an HTTP request to the activity information API and receives event information as a response. This event information includes the event name, date, location, and number of participants. The input is the API key and request URL, and the output is a data frame of event information by date.

[0182] Step 4:

[0183] The server combines the acquired customer visit data, weather information, and event information. Specifically, it combines each data frame using the date as a key to generate a single complete dataset. The input contains each date-specific data frame, and the output contains the combined dataset.

[0184] Step 5:

[0185] The server processes missing and outliers in the merged dataset by imputing missing values ​​with zeros or the mean and filtering outliers. The input is the merged dataset and the output is the complete, processed dataset.

[0186] Step 6:

[0187] The server trains a linear regression model using the preprocessed data. Specifically, it sets weather and event information as features and the number of customer visits as targets, and trains the regression model using the Scikit-learn library. The input is a complete dataset, and the output is a trained model.

[0188] Step 7:

[0189] The server retrieves future weather and event information and preprocesses it. The future weather and event information is again retrieved from an external service via HTTP request and combined using the date as a key. The input contains future weather and event information, and the output contains a dataset for prediction.

[0190] Step 8:

[0191] The server uses the trained linear regression model to predict future customer visits. It inputs preprocessed future weather and event information into the model to generate a prediction result. The input includes a dataset for prediction, and the output includes a prediction result of future customer visits.

[0192] Step 9:

[0193] The terminal receives the forecast data sent from the server and displays it on the user interface. Specifically, it has the function of visually displaying the predicted number of future customer visits, optimal staffing, and inventory management information. The input includes the forecast result data, and the output includes the information displayed on the user interface.

[0194] Step 10:

[0195] Users make decisions based on the forecast data through their devices. For example, they can adjust staff shift schedules or plan inventory replenishment based on the predicted number of customers. The input includes the information displayed on the device, and the output includes the user's decisions and actions.

[0196] 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.

[0197] This system consists of a server and a user terminal, and is further combined with an emotion engine that recognizes the user's emotions. The server obtains the necessary data from weather forecast services and event information services and performs a series of processes to train the model. The emotion engine also recognizes the user's emotions and adjusts the prediction results based on them. The user terminal provides an interface for the user to check the results.

[0198] Explanation of program processing

[0199] Data collection

[0200] 1. The server first retrieves audit data from the historical audit database, which involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame.

[0201] 2. The server sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. The weather information includes data such as temperature, humidity, and weather type for each date.

[0202] 3. The server also sends an HTTP request to an external event information service to retrieve event information for the specified region and period, including the event name, date, location, and number of participants.

[0203] Data Preprocessing

[0204] 4. The server combines the acquired inspection data, weather information, and event information to generate a unified dataset. Each data frame is merged based on date.

[0205] 5. The server will gracefully handle missing values ​​and outliers in the combined datasets, using imputation such as zeros and mean values.

[0206] Training the model

[0207] 6. The server uses the preprocessed data to train a linear regression model, which uses weather and event information as features and the number of reviews as targets.

[0208] 7. The server stores the trained model for future predictions.

[0209] Prediction of future review volume

[0210] 8. The server retrieves future weather and event information, preprocesses it, and reassembles it into a dataset ready to be fed into the model.

[0211] 9. The server uses the trained linear regression model to predict future audit volume. The predictions are added as new columns to the data frame.

[0212] Emotion recognition and predictive outcome adjustment

[0213] 10. The device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data and recognizes the user's emotions.

[0214] 11. The server receives emotion information from the emotion engine and adjusts the prediction results. For example, if the user is feeling stressed, the server may increase the predicted number of reviews.

[0215] Specific examples

[0216] For example, the server collects data on the number of inspections over the past year, while also collecting weather forecasts and event information for the coming month. This data is used to train a linear regression model to predict the number of inspections for the next month. The prediction results are sent to the user's device, where the user can check them. At that time, the emotion engine analyzes the user's facial expression data and voice instructions in real time and provides feedback on the user's emotional state.

[0217] For example, if the user is tired, the system will display a warning on the screen showing the prediction results and suggest increasing the number of reviews, while if the user is relaxed, the system will display standard prediction results.

[0218] In this way, a system incorporating an emotion engine can provide flexible predictions of the number of cases reviewed that take into account the user's emotional state. This functionality not only improves operational efficiency but also the user experience.

[0219] The processing flow will be explained below.

[0220] Step 1:

[0221] The server connects to the database of past review counts. It uses a database connection library, such as sqlite3 or sqlalchemy, to connect to the database. It issues an SQL query to retrieve past review count data from the database. Specifically, it executes a query such as SELECT FROM review_counts and stores the results in a Pandas data frame.

[0222] Step 2:

[0223] The server sends an HTTP request to an external weather forecast service. Using the requests library, it sends parameters including the API key, location information, and forecast period (number of days) to the weather forecast service's API endpoint (e.g., https: / / api.weatherapi.com / v1 / forecast.json). It then obtains the JSON data returned as a response, extracts the necessary weather information (date, time, temperature, weather type, etc.), and converts it into a data frame.

[0224] Step 3:

[0225] The server also sends an HTTP request to the event information service. Using the requests library, it sends parameters including the API key, location information, and duration to the event information service's API endpoint (e.g., https: / / api.eventapi.com / v1 / events). The necessary information (event name, date and time, location, number of attendees, etc.) is extracted from the JSON data returned as a response and converted into a data frame.

[0226] Step 4:

[0227] The server combines the obtained inspection count data, weather information, and event information. It uses the Pandas merge function to combine each data frame using the date as a key. This operation generates a single combined dataset.

[0228] Step 5:

[0229] The server processes missing values ​​in the integrated dataset. It uses Pandas' fillna function to fill missing values ​​with zeros or the average of the preceding and following days. This process ensures the completeness of the dataset.

[0230] Step 6:

[0231] The server trains a linear regression model using the preprocessed data. Using Scikit-learn's LinearRegression class, it separates the features (weather information, event information) and the target (number of reviews) and trains using the model's fit method. Through this training, the model learns parameters from the given data.

[0232] Step 7:

[0233] The server saves the trained model for future predictions. The model is saved using the joblib or pickle libraries, and stored as a file on the local disk or in a database.

[0234] Step 8:

[0235] The server again obtains the weather and event information needed for future predictions. It also sends an HTTP request to obtain future data and performs preprocessing. The obtained data is again incorporated into the integrated dataset using the date as a key.

[0236] Step 9:

[0237] The server uses the trained model to predict future audit volume. It inputs the preprocessed future data into the model and gets the prediction result. The result is added as a new column in the data frame.

[0238] Step 10:

[0239] The server saves the prediction results in a database or sends them to the user's device. When saving the results in a database, an INSERT operation is performed using an appropriate SQL query. When notifying the user, the server sends the results via email or a push notification service, for example.

[0240] Step 11:

[0241] The device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data in real time and recognizes the user's emotions. For example, it uses machine learning libraries and APIs to perform facial recognition and voice analysis.

[0242] Step 12:

[0243] The server receives emotional information from the emotion engine and adjusts the prediction results. For example, if the server detects that the user is feeling stressed, it may increase the predicted number of reviews. This adjustment is achieved by using weights and correction values ​​to reflect the emotional information.

[0244] Step 13:

[0245] Users can check the prediction results on their devices. They can view the predicted number of reviews via a dedicated management screen or notification messages, and use the results to adjust staff schedules and make work plans. They can also provide feedback on the prediction results, contributing to improving the system's prediction accuracy and user experience.

[0246] Example 2

[0247] 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."

[0248] Conventional systems for predicting the number of review cases make predictions based on past data and external information, but they have the problem of being unable to make flexible predictions that take into account the emotional state of the user. In particular, ignoring the impact of user emotions on prediction results can lead to problems such as reduced operational efficiency and a lower user experience.

[0249] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring information on the number of processed cases from a past information acquisition device, means for acquiring weather information from an external weather information acquisition device, means for acquiring event information from an external event information acquisition device, means for preprocessing the acquired information, means for training a prediction model using the preprocessed information, means for predicting the number of processed cases in the future, means for recognizing emotions using an emotion recognition device, and means for adjusting the prediction result based on the recognized emotion information. This makes it possible to reflect the user's emotions in the prediction result, which is expected to improve the user experience and business efficiency.

[0250] The "past information acquisition device" is a means for acquiring necessary information from a database or data storage that stores past information on the number of processed cases.

[0251] The "external weather information acquisition device" is a means for acquiring weather information for a specified area or period from a weather information providing service on the Internet.

[0252] An "external event information acquisition device" is a means for acquiring event information for a specified area or period from an event information providing service on the Internet.

[0253] "Preprocessing" is the operation of integrating the acquired information, appropriately handling missing values ​​and outliers, and converting it into a format suitable for training a predictive model.

[0254] A "predictive model" is a mathematical or statistical model used to predict future transaction volumes based on past and external information.

[0255] An "emotion recognition device" is a means for analyzing a user's facial expressions, voice, character input, etc. to recognize the user's emotional state.

[0256] "Emotion information" is data relating to the user's emotional state obtained by an emotion recognition device.

[0257] "Adjusting the predicted result" is an operation for correcting or amending the predicted number of transactions in accordance with the emotion based on the recognized emotion information.

[0258] The system of the present invention includes a past information acquisition device, an external weather information acquisition device, an external event information acquisition device, a preprocessing device, a prediction model device, an emotion recognition device, and a prediction result adjustment device. Each device and its specific operation are described below.

[0259] First, the server uses the past information acquisition device to acquire the past processing count information from the database. To do this, for example, it uses the Python pandas library to issue an SQL query and acquire the data as a data frame.

[0260] Next, the server uses an external weather information acquisition device to send an HTTP request to the weather information service to obtain weather information for the specified area and period. This operation uses the requests library to obtain information such as temperature, humidity, and weather type.

[0261] The server then uses an external event information acquisition device to send an HTTP request to the event information service to obtain event information for the specified region and period. Again, the requests library is used to collect data such as the event name, date, location, and number of participants.

[0262] The acquired information is preprocessed by the server's preprocessing unit. Specifically, the data frames are merged based on the date using the pandas merge function to integrate the dataset. Missing and outlier values ​​are imputed using the pandas fillna method.

[0263] Next, based on the preprocessed data, the server trains a predictive model using the predictive modeler. Here, a linear regression model is trained using the LinearRegression class of the scikit-learn library. The trained model is saved using the joblib library for future predictions.

[0264] The server retrieves future weather and event information again, preprocesses it, and then uses the trained predictive model to predict future transaction volumes. The prediction results are added as new columns to a pandas data frame.

[0265] Furthermore, the user device runs an emotion recognition device using the user's facial expressions, voice, and text input. Here, the OpenCV library and speech recognition library are used to acquire data in real time and recognize the user's emotions. The recognized emotion information is reflected in the prediction results by the prediction result adjustment device on the server. For example, if the user is feeling stressed, the predicted number of transactions is adjusted to increase.

[0266] Specific examples

[0267] For example, the server retrieves transaction volume information from the database for the previous year, then collects weather and event information for the coming month. Based on this information, a linear regression model is trained using the LinearRegression class in scikit-learn to predict transaction volume for the next month. The results are then sent to the user's device, where the user can view the predictions.

[0268] If a user is tired while using the device, the device will perform facial recognition and voice analysis, and the emotion recognition device will detect the emotion "tired." This information will be sent to the server, and the prediction results will be adjusted. For example, a warning based on the prediction results may be displayed, suggesting an increase in the number of reviews.

[0269] Prompt Sentence Examples

[0270] "Please build a system that predicts the number of cases to be processed next month based on weather and event information."

[0271] "Please add a feature to adjust the predicted transaction volume results based on the user's emotional state."

[0272] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0273] Step 1:

[0274] The server uses the past information acquisition device to obtain information on the number of past processed items from the database. Database connection information and an SQL query are used as input to obtain information on the number of past processed items from the database. Specifically, an SQL query is issued using the Python pandas library, and the results are obtained as a data frame. The output is a data frame containing information on the number of past processed items.

[0275] Step 2:

[0276] The server uses an external weather information acquisition device to send an HTTP request to the weather information service to obtain weather information for the specified region and period. The region, period, and API key are used as input. The API is accessed through the requests library to obtain weather information. The output is data including information such as temperature, humidity, and weather type for each date.

[0277] Step 3:

[0278] The server uses an external event information acquisition device to send an HTTP request to the event information service to obtain event information for the specified region and period. This also uses the region and period information and the API key as input. The requests library is used to access the API and obtain the event information. The output includes information such as the event name, date, location, and number of participants.

[0279] Step 4:

[0280] The server's preprocessing unit combines the acquired transaction count information, weather information, and event information to generate an integrated dataset. It uses a data frame of past transaction count information, weather information data, and event information data as input. It uses the pandas merge function to merge these data frames based on the date. The output is an integrated dataset.

[0281] Step 5:

[0282] The server uses a preprocessor to appropriately handle missing values ​​and outliers in the combined datasets. The input is the combined dataset, and missing and outliers are filled with zeros or the mean value using the pandas fillna method. The output is a dataset with missing and outliers processed.

[0283] Step 6:

[0284] The server trains a predictive model using the preprocessed data. The input is the preprocessed dataset. Using the LinearRegression class from the scikit-learn library, it sets weather information and event information as features and the number of items to be processed as the target. Specifically, it uses the fit method of the model to train the model. The output is the trained predictive model.

[0285] Step 7:

[0286] The server saves the trained predictive model. The input is the trained predictive model. It uses the joblib library to save the model to the file system. The output is the saved model file.

[0287] Step 8:

[0288] The server again retrieves and preprocesses future weather and event information. The input is a future weather information API request and an event information API request. Again, it uses the requests library to retrieve the information and perform preprocessing operations. The output is a dataset containing preprocessed future weather and event information.

[0289] Step 9:

[0290] The server uses the trained prediction model to predict future transaction volumes. The input is preprocessed future weather and event information. The predict method of the prediction model is used to predict future transaction volumes, and the prediction results are added as new columns in a pandas data frame. The output is a dataset with predicted future transaction volumes.

[0291] Step 10:

[0292] The user device runs an emotion recognition device using facial expressions, voice, and text input. The inputs are the user's facial expression data, voice data, and text input data. Using the OpenCV library and voice recognition library, emotions are recognized in real time. The output is the recognized emotion information.

[0293] Step 11:

[0294] The server receives emotion information from the emotion recognition device and adjusts the prediction results. The inputs are the predicted number of transactions and emotion information. For example, if the user is feeling stressed, the server makes adjustments such as increasing the prediction value. The emotion information is reflected in the prediction results and the final prediction results are output.

[0295] (Application example 2)

[0296] 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."

[0297] Conventional prediction systems often make predictions without considering the user's emotions, which often results in insufficient consideration of the user experience. Furthermore, there is a need for systems that can recognize the user's emotional state and respond flexibly based on that. However, conventional technologies lack prediction systems that incorporate emotion recognition, and there is also a lack of means to effectively utilize the prediction results.

[0298] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0299] In this invention, the server includes means for acquiring data on the number of audits from a past database, means for acquiring weather information from an external weather forecast service, means for acquiring event information from an external event information service, means for preprocessing the acquired data, means for training a prediction model using the preprocessed data, means for predicting the number of audits in the future, means for recognizing user emotions and adjusting the prediction results, and means for presenting individually optimized products and services. This enables flexible responses that take user emotions into consideration, makes it possible to more effectively utilize the prediction results, and provides a high-quality user experience.

[0300] "Data on number of examinations" refers to data that includes quantities and statistical information related to past examinations.

[0301] "Weather information" refers to data such as temperature, humidity, and weather type obtained from an external weather forecast service.

[0302] "Event information" refers to data such as the event name, date, venue, and number of participants obtained from an external event information service.

[0303] "Preprocessing" refers to the process of integrating acquired data, treating missing values, and treating outliers.

[0304] A "predictive model" is a mathematical model that is trained based on preprocessed data and is used to predict future numbers of reviews, etc.

[0305] "Emotion recognition" is a technology that analyzes a user's emotions from their facial expressions, voice, text input, etc.

[0306] "Presenting individually optimized products and services" means suggesting optimal products and services in real time based on the user's emotional state and acquired data.

[0307] This invention is a system that is composed of a server and a user terminal, and further combines an emotion engine that recognizes the user's emotions. The server is connected to the Internet and performs the following specific functions.

[0308] First, the server retrieves the historical audit count data from the database. This involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame. The server then sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. It also sends an HTTP request to an external event information service to retrieve event information for the specified region and time period. This data is then consolidated on the server, with missing values ​​and outliers being handled appropriately.

[0309] The server then uses the preprocessed data to train a predictive model, specifically a linear regression model, with weather and event information as features and the number of audits as targets. The trained model is stored on the server for future predictions.

[0310] Meanwhile, the user device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data and recognizes the user's emotions. When the user accesses the virtual store, the emotion engine performs real-time analysis and provides feedback to the server to adjust the prediction results.

[0311] For example, consider a scenario in which a user is shopping online. The user's device uses a smartphone or head-mounted display (HMD) to display products that the user may be interested in. Based on the user's purchasing history, weather information, event information, and emotional data, the server suggests optimal products and promotions in real time. If the user is feeling stressed, the number and variety of suggested products can be adjusted, improving the user experience.

[0312] Possible hardware includes a database server, an API server, and user devices (smartphones, smart glasses, HMDs). Software includes Python, Pandas, Scikit-learn, and virtual libraries for emotion recognition (e.g., SentimentEngine). API communication uses libraries such as the requests library.

[0313] As a concrete example, the server acquires data on the number of inspections over the past year and collects weather forecasts and event information for the coming month. This data is used to train a linear regression model to predict the number of inspections for the next month. The prediction results are sent to the user's device, where the user can check them. At that time, the emotion engine analyzes the user's facial expression data and voice instructions in real time and provides feedback on the user's emotional state.

[0314] Prompt Sentence Examples

[0315] Implement an example of a product recommendation system that takes user emotions into account. Recognize emotions in real time based on past purchase history, weather information, and event information, and suggest optimized products and services.

[0316] Required parts:

[0317] 1. Data Collection and Preprocessing

[0318] 2. Training a model based on the data

[0319] 3. Emotion Recognition Engine Integration

[0320] 4. Real-time optimized recommendations

[0321] Please implement a concrete example in Python.

[0322] By implementing a system based on this prompt, it becomes possible to make suggestions that take into account the user's emotions in real time, providing a high-quality user experience.

[0323] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0324] Step 1:

[0325] The server retrieves historical audit count data from the database. For input, it uses a database connection library and issues an SQL query. The audit count data retrieved from the database is output as a data frame. This data includes the number of audits by date.

[0326] Step 2:

[0327] The server sends an HTTP request to an external weather forecast service to obtain weather information for a specified region and period. The API endpoint and request parameters are required as input. The obtained weather information is output to the server in JSON format and converted into a data frame using a data analysis library. The weather information includes data such as temperature, humidity, and weather type for each date.

[0328] Step 3:

[0329] The server sends an HTTP request to an external event information service to obtain event information for a specified region and period. The API endpoint and request parameters are required as input. The obtained event information is output to the server in JSON format and converted into a data frame using a data analysis library. The event information includes the event name, date, location, number of attendees, etc.

[0330] Step 4:

[0331] The server preprocesses the acquired data. Specifically, it merges the number of inspections, weather information, and event information based on the date. Each data frame is required as input. The merged data integrates the number of inspections, weather information, and event information for each date.

[0332] Step 5:

[0333] The server gracefully handles missing values ​​and outliers present in the preprocessed data. The input is the merged data. Missing values ​​are imputed with zeros or the mean, and outliers are removed or corrected. The output is a clean dataset.

[0334] Step 6:

[0335] The server trains a predictive model (linear regression model) using the clean data. The inputs include features such as weather and event information, and a target variable called the number of reviews. The Scikit-learn library is used to train the model. The trained model is obtained as the output and stored on the server.

[0336] Step 7:

[0337] The server retrieves future weather and event information and preprocesses it. Similar to steps 2 and 3, it retrieves data from the API and converts it into a data frame. As input, it requires future weather and event information, and obtains preprocessed future data as output.

[0338] Step 8:

[0339] The server uses the trained linear regression model to predict future inspection volume. As input, it uses preprocessed future weather and event information. As output, it obtains a data frame of predicted inspection volume.

[0340] Step 9:

[0341] The user's device runs the emotion engine using facial expressions, voice, and text input. When the user uses the application, the emotion engine analyzes input data from the camera and microphone. The inputs required are the user's facial expression data, voice, and text input. The output is a numerical representation of the user's emotional state, which is sent to the server.

[0342] Step 10:

[0343] The server receives emotion information from the emotion engine and adjusts the prediction results. The inputs are emotion information and the predicted number of reviews. For example, if the user is feeling stressed, the server increases the predicted number of reviews. The output is the adjusted prediction results.

[0344] Step 11:

[0345] The user device presents the optimized products and services to the user. Based on the adjusted prediction results, products and promotions that the user is likely to be interested in are displayed. The adjusted prediction results are required as input, and the products and services are presented in a form that the user can see as output.

[0346] 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.

[0347] 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.

[0348] 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.

[0349] [Second embodiment]

[0350] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0351] 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.

[0352] 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).

[0353] 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.

[0354] 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.

[0355] 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).

[0356] 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.

[0357] 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.

[0358] 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.

[0359] 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.

[0360] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0361] 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."

[0362] This system consists of a server and a user terminal. The server acquires the necessary data from weather forecast services and event information services and performs a series of processes to train the model. The user terminal provides an interface for users to check the results.

[0363] Explanation of program processing

[0364] Data collection

[0365] 1. The server first retrieves audit data from the historical audit database, which involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame.

[0366] 2. The server sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. The weather information includes data such as temperature, humidity, and weather type for each date.

[0367] 3. The server also sends an HTTP request to an external event information service to retrieve event information for the specified region and period, including the event name, date, location, and number of participants.

[0368] Data Preprocessing

[0369] 4. The server combines the acquired inspection data, weather information, and event information to generate a unified dataset. Each data frame is merged based on date.

[0370] 5. The server will gracefully handle missing values ​​and outliers in the combined datasets, using imputation such as zeros and mean values.

[0371] Training the model

[0372] 6. The server uses the preprocessed data to train a linear regression model, which uses weather and event information as features and the number of reviews as targets.

[0373] 7. The server stores the trained model for future predictions.

[0374] Prediction of future review volume

[0375] 8. The server retrieves future weather and event information, preprocesses it, and reassembles it into a dataset ready to be fed into the model.

[0376] 9. The server uses the trained linear regression model to predict future audit volume. The prediction results are added to a data frame and can be saved or sent to a device.

[0377] Specific examples

[0378] For example, to predict next month's number of inspections, the server first retrieves data on the number of inspections over the past year. Then, it collects weather and event information for the same period along the time axis, integrates this data, and generates a training dataset. It then trains a linear regression model to predict the number of inspections based on next month's weather and event information. Finally, users can view the results on their devices.

[0379] Users can use the prediction results to, for example, efficiently allocate staff shift schedules or adjust response resources in advance. In this way, the system improves the accuracy of predictions of the number of examination cases, contributing to operational efficiency.

[0380] The processing flow will be explained below.

[0381] Step 1:

[0382] The server connects to the database of past review counts. It uses a database connection library, such as sqlite3 or sqlalchemy, to connect to the database. It issues an SQL query to retrieve past review count data from the database. Specifically, it executes a query such as SELECT FROM review_counts and stores the results in a Pandas data frame.

[0383] Step 2:

[0384] The server sends an HTTP request to the weather forecast service. Using the requests library, it sends parameters including the API key, location information, and forecast period (number of days) to the weather forecast service's API endpoint (e.g., https: / / api.weatherapi.com / v1 / forecast.json). It then obtains the JSON data returned as a response, extracts the necessary weather information (date, time, temperature, weather type, etc.), and converts it into a data frame.

[0385] Step 3:

[0386] The server also sends an HTTP request to the event information service. Using the requests library, it sends parameters including the API key, location information, and duration to the event information service's API endpoint (e.g., https: / / api.eventapi.com / v1 / events). The necessary information (event name, date and time, location, number of attendees, etc.) is extracted from the JSON data returned as a response and converted into a data frame.

[0387] Step 4:

[0388] The server combines the obtained inspection count data, weather information, and event information. It uses the Pandas merge function to combine each data frame using the date as a key. This operation generates a single combined dataset.

[0389] Step 5:

[0390] The server processes missing values ​​in the integrated dataset. It uses Pandas' fillna function to fill missing values ​​with zeros or the average of the preceding and following days. This process ensures the completeness of the dataset.

[0391] Step 6:

[0392] The server trains a linear regression model using the preprocessed data. Using Scikit-learn's LinearRegression class, it separates the features (weather information, event information) and the target (number of reviews) and trains using the model's fit method. Through this training, the model learns parameters from the given data.

[0393] Step 7:

[0394] The server saves the trained model for future predictions. The model is saved using the joblib or pickle libraries, and stored as a file on the local disk or in a database.

[0395] Step 8:

[0396] The server again obtains the weather and event information needed for future predictions. It also sends an HTTP request to obtain future data and performs preprocessing. The obtained data is again incorporated into the integrated dataset using the date as a key.

[0397] Step 9:

[0398] The server uses the trained model to predict future audit volume. It inputs the preprocessed future data into the model and gets the prediction result. The result is added as a new column in the data frame.

[0399] Step 10:

[0400] The server saves the prediction results in a database or sends them to the user's device. When saving the results in a database, an INSERT operation is performed using an appropriate SQL query. When notifying the user, the server sends the results via email or a push notification service, for example.

[0401] Step 11:

[0402] Users can check the forecast results on their devices, view the predicted number of reviews via a dedicated management screen or notification messages, and adjust staff schedules and create work plans based on the results.

[0403] Example 1

[0404] 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."

[0405] Conventional systems had low accuracy in predicting the number of screening cases, making it difficult to improve work efficiency. Furthermore, a series of processes, such as collecting data from external services, preprocessing the data, and training the prediction model, were often performed manually, resulting in a lot of work and prone to errors. Therefore, a more accurate and automated prediction system was needed.

[0406] 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.

[0407] In this invention, the server includes a means for acquiring case count data from a past database, a means for acquiring weather data from an external information service, and a means for acquiring event information from an external information service. This enables the acquired data to be automatically integrated and preprocessed. Furthermore, the preprocessed data can be used to train a prediction model, and future case counts can be predicted with high accuracy based on the prediction results. This improves business efficiency and accuracy.

[0408] The "past database" is a data storage that holds the number of past reviews and related data.

[0409] "Case data" is data that includes information regarding the number of reviews conducted within a particular period of time.

[0410] An "external information service" is an external service that the server accesses via the Internet and that provides weather-related data and forecasts.

[0411] "Weather data" is data that includes information about the weather in a particular region or time period, such as temperature, humidity, and type of weather.

[0412] An "external information service" is an external service that the server accesses via the Internet and provides information about events and other activities.

[0413] "Event information" is data that includes details related to an event, such as the event name, date, location, and number of participants.

[0414] "Preprocessing means" refers to a set of operations or techniques used to process acquired data by cleansing, filtering, integrating, etc., and converting it into a form suitable for training a predictive model.

[0415] "Means of integration" refers to a method of combining data obtained from different sources using dates as a key to create a single unified data set.

[0416] A "predictive model" is a mathematical or statistical model trained to predict future values ​​based on past data.

[0417] A "regression analysis model" is a statistical method for analyzing the relationship between independent variables and dependent variables and making predictions.

[0418] "Future volume" refers to the number of audits that will occur under specified conditions in a projected future period.

[0419] A "server" is a computer system that executes a series of processes such as data acquisition, preprocessing, model training, and prediction calculation.

[0420] The present invention is a system that obtains event count data from a past database, weather data and event information from external information services, and predicts future event counts based on this data. The system is composed of a server and a user terminal.

[0421] Server configuration and functions

[0422] 1. Data Acquisition

[0423] The server first retrieves the historical data from the database. This is done using a database connection library and an SQL query. In Python, the pandas read_sql_query function is used to retrieve the data in data frame format.

[0424] Next, the server retrieves weather data from an external information service (e.g., a weather forecast service API) by sending an HTTP request using the requests library and parsing the resulting data in JSON format.

[0425] Additionally, the server obtains event information from an external event information service (e.g., an event information API). Again, it uses the requests library to send an HTTP request and parses the returned data in JSON format.

[0426] 2. Data Preprocessing

[0427] The server converts the acquired count data, weather data, and event information into respective data frames and combines them based on the date. This combining process uses the pandas merge function. After combining, the data set is filled in with missing values ​​and outliers are processed. For the filling process, the pandas fillna method is used to fill in missing values ​​with zero or the average value.

[0428] 3. Training the model

[0429] The server trains a predictive model (regression analysis model) based on the preprocessed dataset. For training, it uses the LinearRegression class from scikit-learn, learning the model using weather data and event information as features and count data as the target. The trained model is then serialized and saved using the joblib library.

[0430] 4. Future Data Prediction

[0431] The server again retrieves and preprocesses future weather and event data. Based on this new data, it uses the trained regression analysis model to predict future event counts. The prediction results are sent to the user's device or stored locally on the server.

[0432] User terminal configuration and functions

[0433] The user terminal provides an interface for displaying the forecast results sent from the server, allowing the user to improve work efficiency and optimize resource allocation based on these forecast figures.

[0434] Specific examples

[0435] For example, when making a prediction for next month, the server retrieves the number data for the past year from the database. At the same time, it retrieves weather data for the specified period from the weather forecast service API and collects event information from the event information provision API. This data is integrated, missing values ​​are filled in, and a training dataset is generated. The server then trains a regression analysis model based on this dataset and predicts the number of events for next month by inputting future weather data and event information. Finally, users can check these prediction results on their own devices and use them in their work.

[0436] Prompt Sentence Examples

[0437] "Please predict the number of cases next month based on the number of cases over the past year, as well as weather and event information for the specified area."

[0438] As described above, the present invention provides a system that makes highly accurate predictions based on past data and external information, thereby improving business efficiency.

[0439] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0440] Step 1:

[0441] The server first connects to the historical database and retrieves the count data. Specifically, it issues an SQL query using a database connection library. For example, it executes the query "SELECT FROM Review Count WHERE Date BETWEEN '2022-01-01' AND '2022-12-31'" and retrieves the results in data frame format. The input is an SQL query, and the output is count data in data frame format.

[0442] Step 2:

[0443] The server obtains weather data from an external information service. Specifically, it uses the requests library to send an HTTP request to obtain weather information for a specified region and period. The request URL includes parameters such as the API key, region, and period. For example, it is executed as follows: "requests.get('https: / / api.weather.com / v3 / wx / forecast / daily / 5day?apiKey=YOUR_API_KEY&geocode=LAT,LON')". The input is the API request, and the output is weather data in JSON format.

[0444] Step 3:

[0445] The server obtains event information from an external information provider. Again, it uses the requests library to send an HTTP request to obtain event information for a specified region and period. For example, execute it as follows: "requests.get('https: / / api.eventservice.com / v1 / events?apiKey=YOUR_API_KEY&location=LOCATION&startDate=START_DATE&endDate=END_DATE')". The input is the API request, and the output is event information in JSON format.

[0446] Step 4:

[0447] The server integrates the acquired count data, weather data, and event information. To do this, it uses the pandas merge function to combine each data frame based on the date. For example, execute "merged_data = pd.merge(count data, weather data, on='date')". The input is the count data, weather data, and event information in data frame format, and the output is an integrated dataset.

[0448] Step 5:

[0449] The server processes missing values ​​and outliers in the merged dataset. Specifically, it uses the pandas fillna method to fill missing values. Zero or the mean value is used for this filling. For example, execute "merged_data.fillna(0, inplace=True)". The input is the merged dataset, and the output is the imputed dataset.

[0450] Step 6:

[0451] The server trains a regression analysis model using the preprocessed dataset. Using scikit-learn's LinearRegression class, the model is trained using weather data and event information as features and count data as the target. For example, this is done using "from sklearn.linear_model import LinearRegression", "model = LinearRegression()", and "model.fit(features, target)". The input is the preprocessed dataset, and the output is a trained regression analysis model.

[0452] Step 7:

[0453] The server saves the trained model. It serializes and saves the model using the joblib library. For example, run "import joblib" and "joblib.dump(model, 'model.pkl')". The input is the trained regression model, and the output is the saved model file.

[0454] Step 8:

[0455] The server retrieves future weather data and event information again and performs preprocessing. It converts them into a data frame format again and merges them based on the date. For example, execute "future_data = pd.merge(future weather data, future event data, on='date')". The input is future weather data and event information, and the output is a preprocessed future dataset.

[0456] Step 9:

[0457] The server uses the trained regression analysis model to predict future count data. Future weather data and event information are input as features into the model to obtain prediction results. For example, execute "future_predictions = model.predict(future features)". The input is a preprocessed future dataset, and the output is the predicted future count data.

[0458] Step 10:

[0459] The server adds the prediction results to a data frame and sends it to the user's device or saves it in an appropriate database. For example, it can save it as a CSV file using the pandas to_csv method or send it to the user's device through an API. The input is a data frame containing the prediction results, and the output is the saved file or the sent data.

[0460] The above are the processing steps of the program for this system.

[0461] (Application example 1)

[0462] 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."

[0463] In traditional brick-and-mortar store operations, it was difficult to accurately predict the number of customers because operations depended on weather and local event information, making it difficult to optimize staff allocation and inventory management. This also led to staff shortages and overstaffing, hindering efficient store operations. This resulted in increased costs and a decline in service quality.

[0464] 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.

[0465] In this invention, the server includes means for acquiring data from a past database, means for acquiring information from an external weather information service, means for acquiring activity information from an external activity information service, means for preprocessing the acquired data, means for training a prediction model using the preprocessed data, means for predicting the number of customers in the future, means for optimizing staffing using the predicted number of customers in the future, and user interface means for displaying the optimized staffing and inventory management. This makes it possible to accurately predict the number of customers in the future and optimally allocate staff and inventory.

[0466] A "database" is a system designed to efficiently organize and manage information and data.

[0467] "Data acquisition means" refers to the methods and processes used to collect the necessary data.

[0468] "Weather information service" means an external source that provides information about the weather and climate for a particular region or period of time.

[0469] "Activity information services" are external information sources that provide information on local events and activities.

[0470] "Preprocessing" refers to the process of converting and organizing data into a format suitable for analysis and model training.

[0471] A "predictive model" is a computational algorithm for predicting future situations based on past data and factors.

[0472] "Customer traffic forecasting method" means a method or process used to forecast future customer traffic.

[0473] "Staffing optimization measures" are methods and techniques for most efficient staffing.

[0474] A "user interface" is a part of a system that provides a means of input and output that allows a user to directly interact with the system.

[0475] "Store foot traffic" refers to the number of customers who visit a physical store within a specific period of time.

[0476] To implement this invention, a system consisting of a server and a user terminal must be constructed. The server acquires data from a past database and collects information from external weather information services and external activity information services. The user terminal acts as an interface that provides users with forecast results, optimized staffing, and inventory management information.

[0477] ---

[0478] 1. Data Collection

[0479] The server first retrieves customer visit data from a historical database by issuing SQL queries using a database connectivity library and retrieving the results as a data frame, and then retrieves weather and event information using HTTP requests from external weather information services (e.g., OpenWeatherMap API) and activity information services (e.g., Eventful API).

[0480] ---

[0481] 2. Data Preprocessing

[0482] The acquired customer visit data, weather information, and event information are integrated based on the date, and missing or outlier values ​​in the integrated dataset are filled with zeros or the average value.

[0483] ---

[0484] 3. Model training

[0485] A linear regression model is trained using the preprocessed data. This process uses Python's Scikit-learn library. Using weather and event information as features and the number of customer visits as targets, a model is constructed to predict future customer visits.

[0486] ---

[0487] 4. Future Predictions

[0488] Future weather and event information is taken, combined into a preprocessed dataset, and a trained linear regression model is used to predict future store traffic. The prediction results are added to a data frame and saved or sent to the device as a result.

[0489] ---

[0490] 5. User Interface

[0491] The user device displays predicted future customer numbers, optimized staffing, and inventory management information. Staffing plans and inventory management can be visually confirmed using a smartphone app or the robot's display.

[0492] ---

[0493] Specific examples

[0494] For example, to generate a forecast for the next month, the server first retrieves customer visitation data from the past year. Next, it collects weather and event information from the same period along the time axis, integrates this data, and generates a training dataset. It then trains a linear regression model to predict the number of customers based on the weather and event information for the next month. Finally, users can view the results on their devices and, for example, efficiently allocate staff shift schedules or adjust response resources in advance.

[0495] ---

[0496] Prompt Sentence Examples

[0497] "To predict the number of customers coming to Store A next month, we will provide the following information: number of customers visiting the store over the past year, weather information (temperature, humidity, weather type), and event information (event name, date, location, number of participants). Based on this information, please calculate the predicted number of customers coming to the store next month."

[0498] This system is expected to significantly improve the efficiency of physical store operations, as it not only predicts store visits but also enables efficient staff deployment planning and optimized inventory management.

[0499] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0500] Step 1:

[0501] The server retrieves historical customer visit data from a database. Specifically, it uses a database connection library to issue an SQL query and obtains the results as a data frame. This data includes the number of customer visits by date. The input is the database connection information and the SQL query, and the output is a data frame of the number of customer visits by date.

[0502] Step 2:

[0503] The server obtains weather information from an external weather information service. Specifically, it sends an HTTP request to the weather information API and receives weather information as a response. This weather information includes temperature, humidity, and weather type by date. The input is the API key and the request URL, and the output is a data frame of weather information by date.

[0504] Step 3:

[0505] The server obtains event information from an external activity information service. Specifically, it sends an HTTP request to the activity information API and receives event information as a response. This event information includes the event name, date, location, and number of participants. The input is the API key and request URL, and the output is a data frame of event information by date.

[0506] Step 4:

[0507] The server combines the acquired customer visit data, weather information, and event information. Specifically, it combines each data frame using the date as a key to generate a single complete dataset. The input contains each date-specific data frame, and the output contains the combined dataset.

[0508] Step 5:

[0509] The server processes missing and outliers in the merged dataset by imputing missing values ​​with zeros or the mean and filtering outliers. The input is the merged dataset and the output is the complete, processed dataset.

[0510] Step 6:

[0511] The server trains a linear regression model using the preprocessed data. Specifically, it sets weather and event information as features and the number of customer visits as targets, and trains the regression model using the Scikit-learn library. The input is a complete dataset, and the output is a trained model.

[0512] Step 7:

[0513] The server retrieves future weather and event information and preprocesses it. The future weather and event information is again retrieved from an external service via HTTP request and combined using the date as a key. The input contains future weather and event information, and the output contains a dataset for prediction.

[0514] Step 8:

[0515] The server uses the trained linear regression model to predict future customer visits. It inputs preprocessed future weather and event information into the model to generate a prediction result. The input includes a dataset for prediction, and the output includes a prediction result of future customer visits.

[0516] Step 9:

[0517] The terminal receives the forecast data sent from the server and displays it on the user interface. Specifically, it has the function of visually displaying the predicted number of future customer visits, optimal staffing, and inventory management information. The input includes the forecast result data, and the output includes the information displayed on the user interface.

[0518] Step 10:

[0519] Users make decisions based on the forecast data through their devices. For example, they can adjust staff shift schedules or plan inventory replenishment based on the predicted number of customers. The input includes the information displayed on the device, and the output includes the user's decisions and actions.

[0520] 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.

[0521] This system consists of a server and a user terminal, and is further combined with an emotion engine that recognizes the user's emotions. The server obtains the necessary data from weather forecast services and event information services and performs a series of processes to train the model. The emotion engine also recognizes the user's emotions and adjusts the prediction results based on them. The user terminal provides an interface for the user to check the results.

[0522] Explanation of program processing

[0523] Data collection

[0524] 1. The server first retrieves audit data from the historical audit database, which involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame.

[0525] 2. The server sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. The weather information includes data such as temperature, humidity, and weather type for each date.

[0526] 3. The server also sends an HTTP request to an external event information service to retrieve event information for the specified region and period, including the event name, date, location, and number of participants.

[0527] Data Preprocessing

[0528] 4. The server combines the acquired inspection data, weather information, and event information to generate a unified dataset. Each data frame is merged based on date.

[0529] 5. The server will gracefully handle missing values ​​and outliers in the combined datasets, using imputation such as zeros and mean values.

[0530] Training the model

[0531] 6. The server uses the preprocessed data to train a linear regression model, which uses weather and event information as features and the number of reviews as targets.

[0532] 7. The server stores the trained model for future predictions.

[0533] Prediction of future review volume

[0534] 8. The server retrieves future weather and event information, preprocesses it, and reassembles it into a dataset ready to be fed into the model.

[0535] 9. The server uses the trained linear regression model to predict future audit volume. The predictions are added as new columns to the data frame.

[0536] Emotion recognition and predictive outcome adjustment

[0537] 10. The device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data and recognizes the user's emotions.

[0538] 11. The server receives emotion information from the emotion engine and adjusts the prediction results. For example, if the user is feeling stressed, the server may increase the predicted number of reviews.

[0539] Specific examples

[0540] For example, the server collects data on the number of inspections over the past year, while also collecting weather forecasts and event information for the coming month. This data is used to train a linear regression model to predict the number of inspections for the next month. The prediction results are sent to the user's device, where the user can check them. At that time, the emotion engine analyzes the user's facial expression data and voice instructions in real time and provides feedback on the user's emotional state.

[0541] For example, if the user is tired, the system will display a warning on the screen showing the prediction results and suggest increasing the number of reviews, while if the user is relaxed, the system will display standard prediction results.

[0542] In this way, a system incorporating an emotion engine can provide flexible predictions of the number of cases reviewed that take into account the user's emotional state. This functionality not only improves operational efficiency but also the user experience.

[0543] The processing flow will be explained below.

[0544] Step 1:

[0545] The server connects to the database of past review counts. It uses a database connection library, such as sqlite3 or sqlalchemy, to connect to the database. It issues an SQL query to retrieve past review count data from the database. Specifically, it executes a query such as SELECT FROM review_counts and stores the results in a Pandas data frame.

[0546] Step 2:

[0547] The server sends an HTTP request to an external weather forecast service. Using the requests library, it sends parameters including the API key, location information, and forecast period (number of days) to the weather forecast service's API endpoint (e.g., https: / / api.weatherapi.com / v1 / forecast.json). It then obtains the JSON data returned as a response, extracts the necessary weather information (date, time, temperature, weather type, etc.), and converts it into a data frame.

[0548] Step 3:

[0549] The server also sends an HTTP request to the event information service. Using the requests library, it sends parameters including the API key, location information, and duration to the event information service's API endpoint (e.g., https: / / api.eventapi.com / v1 / events). The necessary information (event name, date and time, location, number of attendees, etc.) is extracted from the JSON data returned as a response and converted into a data frame.

[0550] Step 4:

[0551] The server combines the obtained inspection count data, weather information, and event information. It uses the Pandas merge function to combine each data frame using the date as a key. This operation generates a single combined dataset.

[0552] Step 5:

[0553] The server processes missing values ​​in the integrated dataset. It uses Pandas' fillna function to fill missing values ​​with zeros or the average of the preceding and following days. This process ensures the completeness of the dataset.

[0554] Step 6:

[0555] The server trains a linear regression model using the preprocessed data. Using Scikit-learn's LinearRegression class, it separates the features (weather information, event information) and the target (number of reviews) and trains using the model's fit method. Through this training, the model learns parameters from the given data.

[0556] Step 7:

[0557] The server saves the trained model for future predictions. The model is saved using the joblib or pickle libraries, and stored as a file on the local disk or in a database.

[0558] Step 8:

[0559] The server again obtains the weather and event information needed for future predictions. It also sends an HTTP request to obtain future data and performs preprocessing. The obtained data is again incorporated into the integrated dataset using the date as a key.

[0560] Step 9:

[0561] The server uses the trained model to predict future audit volume. It inputs the preprocessed future data into the model and gets the prediction result. The result is added as a new column in the data frame.

[0562] Step 10:

[0563] The server saves the prediction results in a database or sends them to the user's device. When saving the results in a database, an INSERT operation is performed using an appropriate SQL query. When notifying the user, the server sends the results via email or a push notification service, for example.

[0564] Step 11:

[0565] The device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data in real time and recognizes the user's emotions. For example, it uses machine learning libraries and APIs to perform facial recognition and voice analysis.

[0566] Step 12:

[0567] The server receives emotional information from the emotion engine and adjusts the prediction results. For example, if the server detects that the user is feeling stressed, it may increase the predicted number of reviews. This adjustment is achieved by using weights and correction values ​​to reflect the emotional information.

[0568] Step 13:

[0569] Users can check the prediction results on their devices. They can view the predicted number of reviews via a dedicated management screen or notification messages, and use the results to adjust staff schedules and make work plans. They can also provide feedback on the prediction results, contributing to improving the system's prediction accuracy and user experience.

[0570] Example 2

[0571] 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."

[0572] Conventional systems for predicting the number of review cases make predictions based on past data and external information, but they have the problem of being unable to make flexible predictions that take into account the emotional state of the user. In particular, ignoring the impact of user emotions on prediction results can lead to problems such as reduced operational efficiency and a lower user experience.

[0573] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring information on the number of processed cases from a past information acquisition device, means for acquiring weather information from an external weather information acquisition device, means for acquiring event information from an external event information acquisition device, means for preprocessing the acquired information, means for training a prediction model using the preprocessed information, means for predicting the number of processed cases in the future, means for recognizing emotions using an emotion recognition device, and means for adjusting the prediction result based on the recognized emotion information. This makes it possible to reflect the user's emotions in the prediction result, which is expected to improve the user experience and business efficiency.

[0574] The "past information acquisition device" is a means for acquiring necessary information from a database or data storage that stores past information on the number of processed cases.

[0575] The "external weather information acquisition device" is a means for acquiring weather information for a specified area or period from a weather information providing service on the Internet.

[0576] An "external event information acquisition device" is a means for acquiring event information for a specified area or period from an event information providing service on the Internet.

[0577] "Preprocessing" is the operation of integrating the acquired information, appropriately handling missing values ​​and outliers, and converting it into a format suitable for training a predictive model.

[0578] A "predictive model" is a mathematical or statistical model used to predict future transaction volumes based on past and external information.

[0579] An "emotion recognition device" is a means for analyzing a user's facial expressions, voice, character input, etc. to recognize the user's emotional state.

[0580] "Emotion information" is data relating to the user's emotional state obtained by an emotion recognition device.

[0581] "Adjusting the predicted result" is an operation for correcting or amending the predicted number of transactions in accordance with the emotion based on the recognized emotion information.

[0582] The system of the present invention includes a past information acquisition device, an external weather information acquisition device, an external event information acquisition device, a preprocessing device, a prediction model device, an emotion recognition device, and a prediction result adjustment device. Each device and its specific operation are described below.

[0583] First, the server uses the past information acquisition device to acquire the past processing count information from the database. To do this, for example, it uses the Python pandas library to issue an SQL query and acquire the data as a data frame.

[0584] Next, the server uses an external weather information acquisition device to send an HTTP request to the weather information service to obtain weather information for the specified area and period. This operation uses the requests library to obtain information such as temperature, humidity, and weather type.

[0585] The server then uses an external event information acquisition device to send an HTTP request to the event information service to obtain event information for the specified region and period. Again, the requests library is used to collect data such as the event name, date, location, and number of participants.

[0586] The acquired information is preprocessed by the server's preprocessing unit. Specifically, the data frames are merged based on the date using the pandas merge function to integrate the dataset. Missing and outlier values ​​are imputed using the pandas fillna method.

[0587] Next, based on the preprocessed data, the server trains a predictive model using the predictive modeler. Here, a linear regression model is trained using the LinearRegression class of the scikit-learn library. The trained model is saved using the joblib library for future predictions.

[0588] The server retrieves future weather and event information again, preprocesses it, and then uses the trained predictive model to predict future transaction volumes. The prediction results are added as new columns to a pandas data frame.

[0589] Furthermore, the user device runs an emotion recognition device using the user's facial expressions, voice, and text input. Here, the OpenCV library and speech recognition library are used to acquire data in real time and recognize the user's emotions. The recognized emotion information is reflected in the prediction results by the prediction result adjustment device on the server. For example, if the user is feeling stressed, the predicted number of transactions is adjusted to increase.

[0590] Specific examples

[0591] For example, the server retrieves transaction volume information from the database for the previous year, then collects weather and event information for the coming month. Based on this information, a linear regression model is trained using the LinearRegression class in scikit-learn to predict transaction volume for the next month. The results are then sent to the user's device, where the user can view the predictions.

[0592] If a user is tired while using the device, the device will perform facial recognition and voice analysis, and the emotion recognition device will detect the emotion "tired." This information will be sent to the server, and the prediction results will be adjusted. For example, a warning based on the prediction results may be displayed, suggesting an increase in the number of reviews.

[0593] Prompt Sentence Examples

[0594] "Please build a system that predicts the number of cases to be processed next month based on weather and event information."

[0595] "Please add a feature to adjust the predicted transaction volume results based on the user's emotional state."

[0596] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0597] Step 1:

[0598] The server uses the past information acquisition device to obtain information on the number of past processed items from the database. Database connection information and an SQL query are used as input to obtain information on the number of past processed items from the database. Specifically, an SQL query is issued using the Python pandas library, and the results are obtained as a data frame. The output is a data frame containing information on the number of past processed items.

[0599] Step 2:

[0600] The server uses an external weather information acquisition device to send an HTTP request to the weather information service to obtain weather information for the specified region and period. The region, period, and API key are used as input. The API is accessed through the requests library to obtain weather information. The output is data including information such as temperature, humidity, and weather type for each date.

[0601] Step 3:

[0602] The server uses an external event information acquisition device to send an HTTP request to the event information service to obtain event information for the specified region and period. This also uses the region and period information and the API key as input. The requests library is used to access the API and obtain the event information. The output includes information such as the event name, date, location, and number of participants.

[0603] Step 4:

[0604] The server's preprocessing unit combines the acquired transaction count information, weather information, and event information to generate an integrated dataset. It uses a data frame of past transaction count information, weather information data, and event information data as input. It uses the pandas merge function to merge these data frames based on the date. The output is an integrated dataset.

[0605] Step 5:

[0606] The server uses a preprocessor to appropriately handle missing values ​​and outliers in the combined datasets. The input is the combined dataset, and missing and outliers are filled with zeros or the mean value using the pandas fillna method. The output is a dataset with missing and outliers processed.

[0607] Step 6:

[0608] The server trains a predictive model using the preprocessed data. The input is the preprocessed dataset. Using the LinearRegression class from the scikit-learn library, it sets weather information and event information as features and the number of items to be processed as the target. Specifically, it uses the fit method of the model to train the model. The output is the trained predictive model.

[0609] Step 7:

[0610] The server saves the trained predictive model. The input is the trained predictive model. It uses the joblib library to save the model to the file system. The output is the saved model file.

[0611] Step 8:

[0612] The server again retrieves and preprocesses future weather and event information. The input is a future weather information API request and an event information API request. Again, it uses the requests library to retrieve the information and perform preprocessing operations. The output is a dataset containing preprocessed future weather and event information.

[0613] Step 9:

[0614] The server uses the trained prediction model to predict future transaction volumes. The input is preprocessed future weather and event information. The predict method of the prediction model is used to predict future transaction volumes, and the prediction results are added as new columns in a pandas data frame. The output is a dataset with predicted future transaction volumes.

[0615] Step 10:

[0616] The user device runs an emotion recognition device using facial expressions, voice, and text input. The inputs are the user's facial expression data, voice data, and text input data. Using the OpenCV library and voice recognition library, emotions are recognized in real time. The output is the recognized emotion information.

[0617] Step 11:

[0618] The server receives emotion information from the emotion recognition device and adjusts the prediction results. The inputs are the predicted number of transactions and emotion information. For example, if the user is feeling stressed, the server makes adjustments such as increasing the prediction value. The emotion information is reflected in the prediction results and the final prediction results are output.

[0619] (Application example 2)

[0620] 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."

[0621] Conventional prediction systems often make predictions without considering the user's emotions, which often results in insufficient consideration of the user experience. Furthermore, there is a need for systems that can recognize the user's emotional state and respond flexibly based on that. However, conventional technologies lack prediction systems that incorporate emotion recognition, and there is also a lack of means to effectively utilize the prediction results.

[0622] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0623] In this invention, the server includes means for acquiring data on the number of audits from a past database, means for acquiring weather information from an external weather forecast service, means for acquiring event information from an external event information service, means for preprocessing the acquired data, means for training a prediction model using the preprocessed data, means for predicting the number of audits in the future, means for recognizing user emotions and adjusting the prediction results, and means for presenting individually optimized products and services. This enables flexible responses that take user emotions into consideration, makes it possible to more effectively utilize the prediction results, and provides a high-quality user experience.

[0624] "Data on number of examinations" refers to data that includes quantities and statistical information related to past examinations.

[0625] "Weather information" refers to data such as temperature, humidity, and weather type obtained from an external weather forecast service.

[0626] "Event information" refers to data such as the event name, date, venue, and number of participants obtained from an external event information service.

[0627] "Preprocessing" refers to the process of integrating acquired data, treating missing values, and treating outliers.

[0628] A "predictive model" is a mathematical model that is trained based on preprocessed data and is used to predict future numbers of reviews, etc.

[0629] "Emotion recognition" is a technology that analyzes a user's emotions from their facial expressions, voice, text input, etc.

[0630] "Presenting individually optimized products and services" means suggesting optimal products and services in real time based on the user's emotional state and acquired data.

[0631] This invention is a system that is composed of a server and a user terminal, and further combines an emotion engine that recognizes the user's emotions. The server is connected to the Internet and performs the following specific functions.

[0632] First, the server retrieves the historical audit count data from the database. This involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame. The server then sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. It also sends an HTTP request to an external event information service to retrieve event information for the specified region and time period. This data is then consolidated on the server, with missing values ​​and outliers being handled appropriately.

[0633] The server then uses the preprocessed data to train a predictive model, specifically a linear regression model, with weather and event information as features and the number of audits as targets. The trained model is stored on the server for future predictions.

[0634] Meanwhile, the user device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data and recognizes the user's emotions. When the user accesses the virtual store, the emotion engine performs real-time analysis and provides feedback to the server to adjust the prediction results.

[0635] For example, consider a scenario in which a user is shopping online. The user's device uses a smartphone or head-mounted display (HMD) to display products that the user may be interested in. Based on the user's purchasing history, weather information, event information, and emotional data, the server suggests optimal products and promotions in real time. If the user is feeling stressed, the number and variety of suggested products can be adjusted, improving the user experience.

[0636] Possible hardware includes a database server, an API server, and user devices (smartphones, smart glasses, HMDs). Software includes Python, Pandas, Scikit-learn, and virtual libraries for emotion recognition (e.g., SentimentEngine). API communication uses libraries such as the requests library.

[0637] As a concrete example, the server acquires data on the number of inspections over the past year and collects weather forecasts and event information for the coming month. This data is used to train a linear regression model to predict the number of inspections for the next month. The prediction results are sent to the user's device, where the user can check them. At that time, the emotion engine analyzes the user's facial expression data and voice instructions in real time and provides feedback on the user's emotional state.

[0638] Prompt Sentence Examples

[0639] Implement an example of a product recommendation system that takes user emotions into account. Recognize emotions in real time based on past purchase history, weather information, and event information, and suggest optimized products and services.

[0640] Required parts:

[0641] 1. Data Collection and Preprocessing

[0642] 2. Training a model based on the data

[0643] 3. Emotion Recognition Engine Integration

[0644] 4. Real-time optimized recommendations

[0645] Please implement a concrete example in Python.

[0646] By implementing a system based on this prompt, it becomes possible to make suggestions that take into account the user's emotions in real time, providing a high-quality user experience.

[0647] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0648] Step 1:

[0649] The server retrieves historical audit count data from the database. For input, it uses a database connection library and issues an SQL query. The audit count data retrieved from the database is output as a data frame. This data includes the number of audits by date.

[0650] Step 2:

[0651] The server sends an HTTP request to an external weather forecast service to obtain weather information for a specified region and period. The API endpoint and request parameters are required as input. The obtained weather information is output to the server in JSON format and converted into a data frame using a data analysis library. The weather information includes data such as temperature, humidity, and weather type for each date.

[0652] Step 3:

[0653] The server sends an HTTP request to an external event information service to obtain event information for a specified region and period. The API endpoint and request parameters are required as input. The obtained event information is output to the server in JSON format and converted into a data frame using a data analysis library. The event information includes the event name, date, location, number of attendees, etc.

[0654] Step 4:

[0655] The server preprocesses the acquired data. Specifically, it merges the number of inspections, weather information, and event information based on the date. Each data frame is required as input. The merged data integrates the number of inspections, weather information, and event information for each date.

[0656] Step 5:

[0657] The server gracefully handles missing values ​​and outliers present in the preprocessed data. The input is the merged data. Missing values ​​are imputed with zeros or the mean, and outliers are removed or corrected. The output is a clean dataset.

[0658] Step 6:

[0659] The server trains a predictive model (linear regression model) using the clean data. The inputs include features such as weather and event information, and a target variable called the number of reviews. The Scikit-learn library is used to train the model. The trained model is obtained as the output and stored on the server.

[0660] Step 7:

[0661] The server retrieves future weather and event information and preprocesses it. Similar to steps 2 and 3, it retrieves data from the API and converts it into a data frame. As input, it requires future weather and event information, and obtains preprocessed future data as output.

[0662] Step 8:

[0663] The server uses the trained linear regression model to predict future inspection volume. As input, it uses preprocessed future weather and event information. As output, it obtains a data frame of predicted inspection volume.

[0664] Step 9:

[0665] The user's device runs the emotion engine using facial expressions, voice, and text input. When the user uses the application, the emotion engine analyzes input data from the camera and microphone. The inputs required are the user's facial expression data, voice, and text input. The output is a numerical representation of the user's emotional state, which is sent to the server.

[0666] Step 10:

[0667] The server receives emotion information from the emotion engine and adjusts the prediction results. The inputs are emotion information and the predicted number of reviews. For example, if the user is feeling stressed, the server increases the predicted number of reviews. The output is the adjusted prediction results.

[0668] Step 11:

[0669] The user device presents the optimized products and services to the user. Based on the adjusted prediction results, products and promotions that the user is likely to be interested in are displayed. The adjusted prediction results are required as input, and the products and services are presented in a form that the user can see as output.

[0670] 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.

[0671] 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.

[0672] 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.

[0673] [Third embodiment]

[0674] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0675] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0676] 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).

[0677] 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.

[0678] 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.

[0679] 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).

[0680] 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.

[0681] 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.

[0682] 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.

[0683] 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.

[0684] 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.

[0685] 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."

[0686] This system consists of a server and a user terminal. The server acquires the necessary data from weather forecast services and event information services and performs a series of processes to train the model. The user terminal provides an interface for users to check the results.

[0687] Explanation of program processing

[0688] Data collection

[0689] 1. The server first retrieves audit data from the historical audit database, which involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame.

[0690] 2. The server sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. The weather information includes data such as temperature, humidity, and weather type for each date.

[0691] 3. The server also sends an HTTP request to an external event information service to retrieve event information for the specified region and period, including the event name, date, location, and number of participants.

[0692] Data Preprocessing

[0693] 4. The server combines the acquired inspection data, weather information, and event information to generate a unified dataset. Each data frame is merged based on date.

[0694] 5. The server will gracefully handle missing values ​​and outliers in the combined datasets, using imputation such as zeros and mean values.

[0695] Training the model

[0696] 6. The server uses the preprocessed data to train a linear regression model, which uses weather and event information as features and the number of reviews as targets.

[0697] 7. The server stores the trained model for future predictions.

[0698] Prediction of future review volume

[0699] 8. The server retrieves future weather and event information, preprocesses it, and reassembles it into a dataset ready to be fed into the model.

[0700] 9. The server uses the trained linear regression model to predict future audit volume. The prediction results are added to a data frame and can be saved or sent to a device.

[0701] Specific examples

[0702] For example, to predict next month's number of inspections, the server first retrieves data on the number of inspections over the past year. Then, it collects weather and event information for the same period along the time axis, integrates this data, and generates a training dataset. It then trains a linear regression model to predict the number of inspections based on next month's weather and event information. Finally, users can view the results on their devices.

[0703] Users can use the prediction results to, for example, efficiently allocate staff shift schedules or adjust response resources in advance. In this way, the system improves the accuracy of predictions of the number of examination cases, contributing to operational efficiency.

[0704] The processing flow will be explained below.

[0705] Step 1:

[0706] The server connects to the database of past review counts. It uses a database connection library, such as sqlite3 or sqlalchemy, to connect to the database. It issues an SQL query to retrieve past review count data from the database. Specifically, it executes a query such as SELECT FROM review_counts and stores the results in a Pandas data frame.

[0707] Step 2:

[0708] The server sends an HTTP request to the weather forecast service. Using the requests library, it sends parameters including the API key, location information, and forecast period (number of days) to the weather forecast service's API endpoint (e.g., https: / / api.weatherapi.com / v1 / forecast.json). It then obtains the JSON data returned as a response, extracts the necessary weather information (date, time, temperature, weather type, etc.), and converts it into a data frame.

[0709] Step 3:

[0710] The server also sends an HTTP request to the event information service. Using the requests library, it sends parameters including the API key, location information, and duration to the event information service's API endpoint (e.g., https: / / api.eventapi.com / v1 / events). The necessary information (event name, date and time, location, number of attendees, etc.) is extracted from the JSON data returned as a response and converted into a data frame.

[0711] Step 4:

[0712] The server combines the obtained inspection count data, weather information, and event information. It uses the Pandas merge function to combine each data frame using the date as a key. This operation generates a single combined dataset.

[0713] Step 5:

[0714] The server processes missing values ​​in the integrated dataset. It uses Pandas' fillna function to fill missing values ​​with zeros or the average of the preceding and following days. This process ensures the completeness of the dataset.

[0715] Step 6:

[0716] The server trains a linear regression model using the preprocessed data. Using Scikit-learn's LinearRegression class, it separates the features (weather information, event information) and the target (number of reviews) and trains using the model's fit method. Through this training, the model learns parameters from the given data.

[0717] Step 7:

[0718] The server saves the trained model for future predictions. The model is saved using the joblib or pickle libraries, and stored as a file on the local disk or in a database.

[0719] Step 8:

[0720] The server again obtains the weather and event information needed for future predictions. It also sends an HTTP request to obtain future data and performs preprocessing. The obtained data is again incorporated into the integrated dataset using the date as a key.

[0721] Step 9:

[0722] The server uses the trained model to predict future audit volume. It inputs the preprocessed future data into the model and gets the prediction result. The result is added as a new column in the data frame.

[0723] Step 10:

[0724] The server saves the prediction results in a database or sends them to the user's device. When saving the results in a database, an INSERT operation is performed using an appropriate SQL query. When notifying the user, the server sends the results via email or a push notification service, for example.

[0725] Step 11:

[0726] Users can check the forecast results on their devices, view the predicted number of reviews via a dedicated management screen or notification messages, and adjust staff schedules and create work plans based on the results.

[0727] Example 1

[0728] 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."

[0729] Conventional systems had low accuracy in predicting the number of screening cases, making it difficult to improve work efficiency. Furthermore, a series of processes, such as collecting data from external services, preprocessing the data, and training the prediction model, were often performed manually, resulting in a lot of work and prone to errors. Therefore, a more accurate and automated prediction system was needed.

[0730] 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.

[0731] In this invention, the server includes a means for acquiring case count data from a past database, a means for acquiring weather data from an external information service, and a means for acquiring event information from an external information service. This enables the acquired data to be automatically integrated and preprocessed. Furthermore, the preprocessed data can be used to train a prediction model, and future case counts can be predicted with high accuracy based on the prediction results. This improves business efficiency and accuracy.

[0732] The "past database" is a data storage that holds the number of past reviews and related data.

[0733] "Case data" is data that includes information regarding the number of reviews conducted within a particular period of time.

[0734] An "external information service" is an external service that the server accesses via the Internet and that provides weather-related data and forecasts.

[0735] "Weather data" is data that includes information about the weather in a particular region or time period, such as temperature, humidity, and type of weather.

[0736] An "external information service" is an external service that the server accesses via the Internet and provides information about events and other activities.

[0737] "Event information" is data that includes details related to an event, such as the event name, date, location, and number of participants.

[0738] "Preprocessing means" refers to a set of operations or techniques used to process acquired data by cleansing, filtering, integrating, etc., and converting it into a form suitable for training a predictive model.

[0739] "Means of integration" refers to a method of combining data obtained from different sources using dates as a key to create a single unified data set.

[0740] A "predictive model" is a mathematical or statistical model trained to predict future values ​​based on past data.

[0741] A "regression analysis model" is a statistical method for analyzing the relationship between independent variables and dependent variables and making predictions.

[0742] "Future volume" refers to the number of audits that will occur under specified conditions in a projected future period.

[0743] A "server" is a computer system that executes a series of processes such as data acquisition, preprocessing, model training, and prediction calculation.

[0744] The present invention is a system that obtains event count data from a past database, weather data and event information from external information services, and predicts future event counts based on this data. The system is composed of a server and a user terminal.

[0745] Server configuration and functions

[0746] 1. Data Acquisition

[0747] The server first retrieves the historical data from the database. This is done using a database connection library and an SQL query. In Python, the pandas read_sql_query function is used to retrieve the data in data frame format.

[0748] Next, the server retrieves weather data from an external information service (e.g., a weather forecast service API) by sending an HTTP request using the requests library and parsing the resulting data in JSON format.

[0749] Additionally, the server obtains event information from an external event information service (e.g., an event information API). Again, it uses the requests library to send an HTTP request and parses the returned data in JSON format.

[0750] 2. Data Preprocessing

[0751] The server converts the acquired count data, weather data, and event information into respective data frames and combines them based on the date. This combining process uses the pandas merge function. After combining, the data set is filled in with missing values ​​and outliers are processed. For the filling process, the pandas fillna method is used to fill in missing values ​​with zero or the average value.

[0752] 3. Training the model

[0753] The server trains a predictive model (regression analysis model) based on the preprocessed dataset. For training, it uses the LinearRegression class from scikit-learn, learning the model using weather data and event information as features and count data as the target. The trained model is then serialized and saved using the joblib library.

[0754] 4. Future Data Prediction

[0755] The server again retrieves and preprocesses future weather and event data. Based on this new data, it uses the trained regression analysis model to predict future event counts. The prediction results are sent to the user's device or stored locally on the server.

[0756] User terminal configuration and functions

[0757] The user terminal provides an interface for displaying the forecast results sent from the server, allowing the user to improve work efficiency and optimize resource allocation based on these forecast figures.

[0758] Specific examples

[0759] For example, when making a prediction for next month, the server retrieves the number data for the past year from the database. At the same time, it retrieves weather data for the specified period from the weather forecast service API and collects event information from the event information provision API. This data is integrated, missing values ​​are filled in, and a training dataset is generated. The server then trains a regression analysis model based on this dataset and predicts the number of events for next month by inputting future weather data and event information. Finally, users can check these prediction results on their own devices and use them in their work.

[0760] Prompt Sentence Examples

[0761] "Please predict the number of cases next month based on the number of cases over the past year, as well as weather and event information for the specified area."

[0762] As described above, the present invention provides a system that makes highly accurate predictions based on past data and external information, thereby improving business efficiency.

[0763] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0764] Step 1:

[0765] The server first connects to the historical database and retrieves the count data. Specifically, it issues an SQL query using a database connection library. For example, it executes the query "SELECT FROM Review Count WHERE Date BETWEEN '2022-01-01' AND '2022-12-31'" and retrieves the results in data frame format. The input is an SQL query, and the output is count data in data frame format.

[0766] Step 2:

[0767] The server obtains weather data from an external information service. Specifically, it uses the requests library to send an HTTP request to obtain weather information for a specified region and period. The request URL includes parameters such as the API key, region, and period. For example, it is executed as follows: "requests.get('https: / / api.weather.com / v3 / wx / forecast / daily / 5day?apiKey=YOUR_API_KEY&geocode=LAT,LON')". The input is the API request, and the output is weather data in JSON format.

[0768] Step 3:

[0769] The server obtains event information from an external information provider. Again, it uses the requests library to send an HTTP request to obtain event information for a specified region and period. For example, execute it as follows: "requests.get('https: / / api.eventservice.com / v1 / events?apiKey=YOUR_API_KEY&location=LOCATION&startDate=START_DATE&endDate=END_DATE')". The input is the API request, and the output is event information in JSON format.

[0770] Step 4:

[0771] The server integrates the acquired count data, weather data, and event information. To do this, it uses the pandas merge function to combine each data frame based on the date. For example, execute "merged_data = pd.merge(count data, weather data, on='date')". The input is the count data, weather data, and event information in data frame format, and the output is an integrated dataset.

[0772] Step 5:

[0773] The server processes missing values ​​and outliers in the merged dataset. Specifically, it uses the pandas fillna method to fill missing values. Zero or the mean value is used for this filling. For example, execute "merged_data.fillna(0, inplace=True)". The input is the merged dataset, and the output is the imputed dataset.

[0774] Step 6:

[0775] The server trains a regression analysis model using the preprocessed dataset. Using scikit-learn's LinearRegression class, the model is trained using weather data and event information as features and count data as the target. For example, this is done using "from sklearn.linear_model import LinearRegression", "model = LinearRegression()", and "model.fit(features, target)". The input is the preprocessed dataset, and the output is a trained regression analysis model.

[0776] Step 7:

[0777] The server saves the trained model. It serializes and saves the model using the joblib library. For example, run "import joblib" and "joblib.dump(model, 'model.pkl')". The input is the trained regression model, and the output is the saved model file.

[0778] Step 8:

[0779] The server retrieves future weather data and event information again and performs preprocessing. It converts them into a data frame format again and merges them based on the date. For example, execute "future_data = pd.merge(future weather data, future event data, on='date')". The input is future weather data and event information, and the output is a preprocessed future dataset.

[0780] Step 9:

[0781] The server uses the trained regression analysis model to predict future count data. Future weather data and event information are input as features into the model to obtain prediction results. For example, execute "future_predictions = model.predict(future features)". The input is a preprocessed future dataset, and the output is the predicted future count data.

[0782] Step 10:

[0783] The server adds the prediction results to a data frame and sends it to the user's device or saves it in an appropriate database. For example, it can save it as a CSV file using the pandas to_csv method or send it to the user's device through an API. The input is a data frame containing the prediction results, and the output is the saved file or the sent data.

[0784] The above are the processing steps of the program for this system.

[0785] (Application example 1)

[0786] 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."

[0787] In traditional brick-and-mortar store operations, it was difficult to accurately predict the number of customers because operations depended on weather and local event information, making it difficult to optimize staff allocation and inventory management. This also led to staff shortages and overstaffing, hindering efficient store operations. This resulted in increased costs and a decline in service quality.

[0788] 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.

[0789] In this invention, the server includes means for acquiring data from a past database, means for acquiring information from an external weather information service, means for acquiring activity information from an external activity information service, means for preprocessing the acquired data, means for training a prediction model using the preprocessed data, means for predicting the number of customers in the future, means for optimizing staffing using the predicted number of customers in the future, and user interface means for displaying the optimized staffing and inventory management. This makes it possible to accurately predict the number of customers in the future and optimally allocate staff and inventory.

[0790] A "database" is a system designed to efficiently organize and manage information and data.

[0791] "Data acquisition means" refers to the methods and processes used to collect the necessary data.

[0792] "Weather information service" means an external source that provides information about the weather and climate for a particular region or period of time.

[0793] "Activity information services" are external information sources that provide information on local events and activities.

[0794] "Preprocessing" refers to the process of converting and organizing data into a format suitable for analysis and model training.

[0795] A "predictive model" is a computational algorithm for predicting future situations based on past data and factors.

[0796] "Customer traffic forecasting method" means a method or process used to forecast future customer traffic.

[0797] "Staffing optimization measures" are methods and techniques for most efficient staffing.

[0798] A "user interface" is a part of a system that provides a means of input and output that allows a user to directly interact with the system.

[0799] "Store foot traffic" refers to the number of customers who visit a physical store within a specific period of time.

[0800] To implement this invention, a system consisting of a server and a user terminal must be constructed. The server acquires data from a past database and collects information from external weather information services and external activity information services. The user terminal acts as an interface that provides users with forecast results, optimized staffing, and inventory management information.

[0801] ---

[0802] 1. Data Collection

[0803] The server first retrieves customer visit data from a historical database by issuing SQL queries using a database connectivity library and retrieving the results as a data frame, and then retrieves weather and event information using HTTP requests from external weather information services (e.g., OpenWeatherMap API) and activity information services (e.g., Eventful API).

[0804] ---

[0805] 2. Data Preprocessing

[0806] The acquired customer visit data, weather information, and event information are integrated based on the date, and missing or outlier values ​​in the integrated dataset are filled with zeros or the average value.

[0807] ---

[0808] 3. Model training

[0809] A linear regression model is trained using the preprocessed data. This process uses Python's Scikit-learn library. Using weather and event information as features and the number of customer visits as targets, a model is constructed to predict future customer visits.

[0810] ---

[0811] 4. Future Predictions

[0812] Future weather and event information is taken, combined into a preprocessed dataset, and a trained linear regression model is used to predict future store traffic. The prediction results are added to a data frame and saved or sent to the device as a result.

[0813] ---

[0814] 5. User Interface

[0815] The user device displays predicted future customer numbers, optimized staffing, and inventory management information. Staffing plans and inventory management can be visually confirmed using a smartphone app or the robot's display.

[0816] ---

[0817] Specific examples

[0818] For example, to generate a forecast for the next month, the server first retrieves customer visitation data from the past year. Next, it collects weather and event information from the same period along the time axis, integrates this data, and generates a training dataset. It then trains a linear regression model to predict the number of customers based on the weather and event information for the next month. Finally, users can view the results on their devices and, for example, efficiently allocate staff shift schedules or adjust response resources in advance.

[0819] ---

[0820] Prompt Sentence Examples

[0821] "To predict the number of customers coming to Store A next month, we will provide the following information: number of customers visiting the store over the past year, weather information (temperature, humidity, weather type), and event information (event name, date, location, number of participants). Based on this information, please calculate the predicted number of customers coming to the store next month."

[0822] This system is expected to significantly improve the efficiency of physical store operations, as it not only predicts store visits but also enables efficient staff deployment planning and optimized inventory management.

[0823] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0824] Step 1:

[0825] The server retrieves historical customer visit data from a database. Specifically, it uses a database connection library to issue an SQL query and obtains the results as a data frame. This data includes the number of customer visits by date. The input is the database connection information and the SQL query, and the output is a data frame of the number of customer visits by date.

[0826] Step 2:

[0827] The server obtains weather information from an external weather information service. Specifically, it sends an HTTP request to the weather information API and receives weather information as a response. This weather information includes temperature, humidity, and weather type by date. The input is the API key and the request URL, and the output is a data frame of weather information by date.

[0828] Step 3:

[0829] The server obtains event information from an external activity information service. Specifically, it sends an HTTP request to the activity information API and receives event information as a response. This event information includes the event name, date, location, and number of participants. The input is the API key and request URL, and the output is a data frame of event information by date.

[0830] Step 4:

[0831] The server combines the acquired customer visit data, weather information, and event information. Specifically, it combines each data frame using the date as a key to generate a single complete dataset. The input contains each date-specific data frame, and the output contains the combined dataset.

[0832] Step 5:

[0833] The server processes missing and outliers in the merged dataset by imputing missing values ​​with zeros or the mean and filtering outliers. The input is the merged dataset and the output is the complete, processed dataset.

[0834] Step 6:

[0835] The server trains a linear regression model using the preprocessed data. Specifically, it sets weather and event information as features and the number of customer visits as targets, and trains the regression model using the Scikit-learn library. The input is a complete dataset, and the output is a trained model.

[0836] Step 7:

[0837] The server retrieves future weather and event information and preprocesses it. The future weather and event information is again retrieved from an external service via HTTP request and combined using the date as a key. The input contains future weather and event information, and the output contains a dataset for prediction.

[0838] Step 8:

[0839] The server uses the trained linear regression model to predict future customer visits. It inputs preprocessed future weather and event information into the model to generate a prediction result. The input includes a dataset for prediction, and the output includes a prediction result of future customer visits.

[0840] Step 9:

[0841] The terminal receives the forecast data sent from the server and displays it on the user interface. Specifically, it has the function of visually displaying the predicted number of future customer visits, optimal staffing, and inventory management information. The input includes the forecast result data, and the output includes the information displayed on the user interface.

[0842] Step 10:

[0843] Users make decisions based on the forecast data through their devices. For example, they can adjust staff shift schedules or plan inventory replenishment based on the predicted number of customers. The input includes the information displayed on the device, and the output includes the user's decisions and actions.

[0844] 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.

[0845] This system consists of a server and a user terminal, and is further combined with an emotion engine that recognizes the user's emotions. The server obtains the necessary data from weather forecast services and event information services and performs a series of processes to train the model. The emotion engine also recognizes the user's emotions and adjusts the prediction results based on them. The user terminal provides an interface for the user to check the results.

[0846] Explanation of program processing

[0847] Data collection

[0848] 1. The server first retrieves audit data from the historical audit database, which involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame.

[0849] 2. The server sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. The weather information includes data such as temperature, humidity, and weather type for each date.

[0850] 3. The server also sends an HTTP request to an external event information service to retrieve event information for the specified region and period, including the event name, date, location, and number of participants.

[0851] Data Preprocessing

[0852] 4. The server combines the acquired inspection data, weather information, and event information to generate a unified dataset. Each data frame is merged based on date.

[0853] 5. The server will gracefully handle missing values ​​and outliers in the combined datasets, using imputation such as zeros and mean values.

[0854] Training the model

[0855] 6. The server uses the preprocessed data to train a linear regression model, which uses weather and event information as features and the number of reviews as targets.

[0856] 7. The server stores the trained model for future predictions.

[0857] Prediction of future review volume

[0858] 8. The server retrieves future weather and event information, preprocesses it, and reassembles it into a dataset ready to be fed into the model.

[0859] 9. The server uses the trained linear regression model to predict future audit volume. The predictions are added as new columns to the data frame.

[0860] Emotion recognition and predictive outcome adjustment

[0861] 10. The device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data and recognizes the user's emotions.

[0862] 11. The server receives emotion information from the emotion engine and adjusts the prediction results. For example, if the user is feeling stressed, the server may increase the predicted number of reviews.

[0863] Specific examples

[0864] For example, the server collects data on the number of inspections over the past year, while also collecting weather forecasts and event information for the coming month. This data is used to train a linear regression model to predict the number of inspections for the next month. The prediction results are sent to the user's device, where the user can check them. At that time, the emotion engine analyzes the user's facial expression data and voice instructions in real time and provides feedback on the user's emotional state.

[0865] For example, if the user is tired, the system will display a warning on the screen showing the prediction results and suggest increasing the number of reviews, while if the user is relaxed, the system will display standard prediction results.

[0866] In this way, a system incorporating an emotion engine can provide flexible predictions of the number of cases reviewed that take into account the user's emotional state. This functionality not only improves operational efficiency but also the user experience.

[0867] The processing flow will be explained below.

[0868] Step 1:

[0869] The server connects to the database of past review counts. It uses a database connection library, such as sqlite3 or sqlalchemy, to connect to the database. It issues an SQL query to retrieve past review count data from the database. Specifically, it executes a query such as SELECT FROM review_counts and stores the results in a Pandas data frame.

[0870] Step 2:

[0871] The server sends an HTTP request to an external weather forecast service. Using the requests library, it sends parameters including the API key, location information, and forecast period (number of days) to the weather forecast service's API endpoint (e.g., https: / / api.weatherapi.com / v1 / forecast.json). It then obtains the JSON data returned as a response, extracts the necessary weather information (date, time, temperature, weather type, etc.), and converts it into a data frame.

[0872] Step 3:

[0873] The server also sends an HTTP request to the event information service. Using the requests library, it sends parameters including the API key, location information, and duration to the event information service's API endpoint (e.g., https: / / api.eventapi.com / v1 / events). The necessary information (event name, date and time, location, number of attendees, etc.) is extracted from the JSON data returned as a response and converted into a data frame.

[0874] Step 4:

[0875] The server combines the obtained inspection count data, weather information, and event information. It uses the Pandas merge function to combine each data frame using the date as a key. This operation generates a single combined dataset.

[0876] Step 5:

[0877] The server processes missing values ​​in the integrated dataset. It uses Pandas' fillna function to fill missing values ​​with zeros or the average of the preceding and following days. This process ensures the completeness of the dataset.

[0878] Step 6:

[0879] The server trains a linear regression model using the preprocessed data. Using Scikit-learn's LinearRegression class, it separates the features (weather information, event information) and the target (number of reviews) and trains using the model's fit method. Through this training, the model learns parameters from the given data.

[0880] Step 7:

[0881] The server saves the trained model for future predictions. The model is saved using the joblib or pickle libraries, and stored as a file on the local disk or in a database.

[0882] Step 8:

[0883] The server again obtains the weather and event information needed for future predictions. It also sends an HTTP request to obtain future data and performs preprocessing. The obtained data is again incorporated into the integrated dataset using the date as a key.

[0884] Step 9:

[0885] The server uses the trained model to predict future audit volume. It inputs the preprocessed future data into the model and gets the prediction result. The result is added as a new column in the data frame.

[0886] Step 10:

[0887] The server saves the prediction results in a database or sends them to the user's device. When saving the results in a database, an INSERT operation is performed using an appropriate SQL query. When notifying the user, the server sends the results via email or a push notification service, for example.

[0888] Step 11:

[0889] The device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data in real time and recognizes the user's emotions. For example, it uses machine learning libraries and APIs to perform facial recognition and voice analysis.

[0890] Step 12:

[0891] The server receives emotional information from the emotion engine and adjusts the prediction results. For example, if the server detects that the user is feeling stressed, it may increase the predicted number of reviews. This adjustment is achieved by using weights and correction values ​​to reflect the emotional information.

[0892] Step 13:

[0893] Users can check the prediction results on their devices. They can view the predicted number of reviews via a dedicated management screen or notification messages, and use the results to adjust staff schedules and make work plans. They can also provide feedback on the prediction results, contributing to improving the system's prediction accuracy and user experience.

[0894] Example 2

[0895] 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."

[0896] Conventional systems for predicting the number of review cases make predictions based on past data and external information, but they have the problem of being unable to make flexible predictions that take into account the emotional state of the user. In particular, ignoring the impact of user emotions on prediction results can lead to problems such as reduced operational efficiency and a lower user experience.

[0897] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring information on the number of processed cases from a past information acquisition device, means for acquiring weather information from an external weather information acquisition device, means for acquiring event information from an external event information acquisition device, means for preprocessing the acquired information, means for training a prediction model using the preprocessed information, means for predicting the number of processed cases in the future, means for recognizing emotions using an emotion recognition device, and means for adjusting the prediction result based on the recognized emotion information. This makes it possible to reflect the user's emotions in the prediction result, which is expected to improve the user experience and business efficiency.

[0898] The "past information acquisition device" is a means for acquiring necessary information from a database or data storage that stores past information on the number of processed cases.

[0899] The "external weather information acquisition device" is a means for acquiring weather information for a specified area or period from a weather information providing service on the Internet.

[0900] An "external event information acquisition device" is a means for acquiring event information for a specified area or period from an event information providing service on the Internet.

[0901] "Preprocessing" is the operation of integrating the acquired information, appropriately handling missing values ​​and outliers, and converting it into a format suitable for training a predictive model.

[0902] A "predictive model" is a mathematical or statistical model used to predict future transaction volumes based on past and external information.

[0903] An "emotion recognition device" is a means for analyzing a user's facial expressions, voice, character input, etc. to recognize the user's emotional state.

[0904] "Emotion information" is data relating to the user's emotional state obtained by an emotion recognition device.

[0905] "Adjusting the predicted result" is an operation for correcting or amending the predicted number of transactions in accordance with the emotion based on the recognized emotion information.

[0906] The system of the present invention includes a past information acquisition device, an external weather information acquisition device, an external event information acquisition device, a preprocessing device, a prediction model device, an emotion recognition device, and a prediction result adjustment device. Each device and its specific operation are described below.

[0907] First, the server uses the past information acquisition device to acquire the past processing count information from the database. To do this, for example, it uses the Python pandas library to issue an SQL query and acquire the data as a data frame.

[0908] Next, the server uses an external weather information acquisition device to send an HTTP request to the weather information service to obtain weather information for the specified area and period. This operation uses the requests library to obtain information such as temperature, humidity, and weather type.

[0909] The server then uses an external event information acquisition device to send an HTTP request to the event information service to obtain event information for the specified region and period. Again, the requests library is used to collect data such as the event name, date, location, and number of participants.

[0910] The acquired information is preprocessed by the server's preprocessing unit. Specifically, the data frames are merged based on the date using the pandas merge function to integrate the dataset. Missing and outlier values ​​are imputed using the pandas fillna method.

[0911] Next, based on the preprocessed data, the server trains a predictive model using the predictive modeler. Here, a linear regression model is trained using the LinearRegression class of the scikit-learn library. The trained model is saved using the joblib library for future predictions.

[0912] The server retrieves future weather and event information again, preprocesses it, and then uses the trained predictive model to predict future transaction volumes. The prediction results are added as new columns to a pandas data frame.

[0913] Furthermore, the user device runs an emotion recognition device using the user's facial expressions, voice, and text input. Here, the OpenCV library and speech recognition library are used to acquire data in real time and recognize the user's emotions. The recognized emotion information is reflected in the prediction results by the prediction result adjustment device on the server. For example, if the user is feeling stressed, the predicted number of transactions is adjusted to increase.

[0914] Specific examples

[0915] For example, the server retrieves transaction volume information from the database for the previous year, then collects weather and event information for the coming month. Based on this information, a linear regression model is trained using the LinearRegression class in scikit-learn to predict transaction volume for the next month. The results are then sent to the user's device, where the user can view the predictions.

[0916] If a user is tired while using the device, the device will perform facial recognition and voice analysis, and the emotion recognition device will detect the emotion "tired." This information will be sent to the server, and the prediction results will be adjusted. For example, a warning based on the prediction results may be displayed, suggesting an increase in the number of reviews.

[0917] Prompt Sentence Examples

[0918] "Please build a system that predicts the number of cases to be processed next month based on weather and event information."

[0919] "Please add a feature to adjust the predicted transaction volume results based on the user's emotional state."

[0920] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0921] Step 1:

[0922] The server uses the past information acquisition device to obtain information on the number of past processed items from the database. Database connection information and an SQL query are used as input to obtain information on the number of past processed items from the database. Specifically, an SQL query is issued using the Python pandas library, and the results are obtained as a data frame. The output is a data frame containing information on the number of past processed items.

[0923] Step 2:

[0924] The server uses an external weather information acquisition device to send an HTTP request to the weather information service to obtain weather information for the specified region and period. The region, period, and API key are used as input. The API is accessed through the requests library to obtain weather information. The output is data including information such as temperature, humidity, and weather type for each date.

[0925] Step 3:

[0926] The server uses an external event information acquisition device to send an HTTP request to the event information service to obtain event information for the specified region and period. This also uses the region and period information and the API key as input. The requests library is used to access the API and obtain the event information. The output includes information such as the event name, date, location, and number of participants.

[0927] Step 4:

[0928] The server's preprocessing unit combines the acquired transaction count information, weather information, and event information to generate an integrated dataset. It uses a data frame of past transaction count information, weather information data, and event information data as input. It uses the pandas merge function to merge these data frames based on the date. The output is an integrated dataset.

[0929] Step 5:

[0930] The server uses a preprocessor to appropriately handle missing values ​​and outliers in the combined datasets. The input is the combined dataset, and missing and outliers are filled with zeros or the mean value using the pandas fillna method. The output is a dataset with missing and outliers processed.

[0931] Step 6:

[0932] The server trains a predictive model using the preprocessed data. The input is the preprocessed dataset. Using the LinearRegression class from the scikit-learn library, it sets weather information and event information as features and the number of items to be processed as the target. Specifically, it uses the fit method of the model to train the model. The output is the trained predictive model.

[0933] Step 7:

[0934] The server saves the trained predictive model. The input is the trained predictive model. It uses the joblib library to save the model to the file system. The output is the saved model file.

[0935] Step 8:

[0936] The server again retrieves and preprocesses future weather and event information. The input is a future weather information API request and an event information API request. Again, it uses the requests library to retrieve the information and perform preprocessing operations. The output is a dataset containing preprocessed future weather and event information.

[0937] Step 9:

[0938] The server uses the trained prediction model to predict future transaction volumes. The input is preprocessed future weather and event information. The predict method of the prediction model is used to predict future transaction volumes, and the prediction results are added as new columns in a pandas data frame. The output is a dataset with predicted future transaction volumes.

[0939] Step 10:

[0940] The user device runs an emotion recognition device using facial expressions, voice, and text input. The inputs are the user's facial expression data, voice data, and text input data. Using the OpenCV library and voice recognition library, emotions are recognized in real time. The output is the recognized emotion information.

[0941] Step 11:

[0942] The server receives emotion information from the emotion recognition device and adjusts the prediction results. The inputs are the predicted number of transactions and emotion information. For example, if the user is feeling stressed, the server makes adjustments such as increasing the prediction value. The emotion information is reflected in the prediction results and the final prediction results are output.

[0943] (Application example 2)

[0944] 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."

[0945] Conventional prediction systems often make predictions without considering the user's emotions, which often results in insufficient consideration of the user experience. Furthermore, there is a need for systems that can recognize the user's emotional state and respond flexibly based on that. However, conventional technologies lack prediction systems that incorporate emotion recognition, and there is also a lack of means to effectively utilize the prediction results.

[0946] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0947] In this invention, the server includes means for acquiring data on the number of audits from a past database, means for acquiring weather information from an external weather forecast service, means for acquiring event information from an external event information service, means for preprocessing the acquired data, means for training a prediction model using the preprocessed data, means for predicting the number of audits in the future, means for recognizing user emotions and adjusting the prediction results, and means for presenting individually optimized products and services. This enables flexible responses that take user emotions into consideration, makes it possible to more effectively utilize the prediction results, and provides a high-quality user experience.

[0948] "Data on number of examinations" refers to data that includes quantities and statistical information related to past examinations.

[0949] "Weather information" refers to data such as temperature, humidity, and weather type obtained from an external weather forecast service.

[0950] "Event information" refers to data such as the event name, date, venue, and number of participants obtained from an external event information service.

[0951] "Preprocessing" refers to the process of integrating acquired data, treating missing values, and treating outliers.

[0952] A "predictive model" is a mathematical model that is trained based on preprocessed data and is used to predict future numbers of reviews, etc.

[0953] "Emotion recognition" is a technology that analyzes a user's emotions from their facial expressions, voice, text input, etc.

[0954] "Presenting individually optimized products and services" means suggesting optimal products and services in real time based on the user's emotional state and acquired data.

[0955] This invention is a system that is composed of a server and a user terminal, and further combines an emotion engine that recognizes the user's emotions. The server is connected to the Internet and performs the following specific functions.

[0956] First, the server retrieves the historical audit count data from the database. This involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame. The server then sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. It also sends an HTTP request to an external event information service to retrieve event information for the specified region and time period. This data is then consolidated on the server, with missing values ​​and outliers being handled appropriately.

[0957] The server then uses the preprocessed data to train a predictive model, specifically a linear regression model, with weather and event information as features and the number of audits as targets. The trained model is stored on the server for future predictions.

[0958] Meanwhile, the user device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data and recognizes the user's emotions. When the user accesses the virtual store, the emotion engine performs real-time analysis and provides feedback to the server to adjust the prediction results.

[0959] For example, consider a scenario in which a user is shopping online. The user's device uses a smartphone or head-mounted display (HMD) to display products that the user may be interested in. Based on the user's purchasing history, weather information, event information, and emotional data, the server suggests optimal products and promotions in real time. If the user is feeling stressed, the number and variety of suggested products can be adjusted, improving the user experience.

[0960] Possible hardware includes a database server, an API server, and user devices (smartphones, smart glasses, HMDs). Software includes Python, Pandas, Scikit-learn, and virtual libraries for emotion recognition (e.g., SentimentEngine). API communication uses libraries such as the requests library.

[0961] As a concrete example, the server acquires data on the number of inspections over the past year and collects weather forecasts and event information for the coming month. This data is used to train a linear regression model to predict the number of inspections for the next month. The prediction results are sent to the user's device, where the user can check them. At that time, the emotion engine analyzes the user's facial expression data and voice instructions in real time and provides feedback on the user's emotional state.

[0962] Prompt Sentence Examples

[0963] Implement an example of a product recommendation system that takes user emotions into account. Recognize emotions in real time based on past purchase history, weather information, and event information, and suggest optimized products and services.

[0964] Required parts:

[0965] 1. Data Collection and Preprocessing

[0966] 2. Training a model based on the data

[0967] 3. Emotion Recognition Engine Integration

[0968] 4. Real-time optimized recommendations

[0969] Please implement a concrete example in Python.

[0970] By implementing a system based on this prompt, it becomes possible to make suggestions that take into account the user's emotions in real time, providing a high-quality user experience.

[0971] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0972] Step 1:

[0973] The server retrieves historical audit count data from the database. For input, it uses a database connection library and issues an SQL query. The audit count data retrieved from the database is output as a data frame. This data includes the number of audits by date.

[0974] Step 2:

[0975] The server sends an HTTP request to an external weather forecast service to obtain weather information for a specified region and period. The API endpoint and request parameters are required as input. The obtained weather information is output to the server in JSON format and converted into a data frame using a data analysis library. The weather information includes data such as temperature, humidity, and weather type for each date.

[0976] Step 3:

[0977] The server sends an HTTP request to an external event information service to obtain event information for a specified region and period. The API endpoint and request parameters are required as input. The obtained event information is output to the server in JSON format and converted into a data frame using a data analysis library. The event information includes the event name, date, location, number of attendees, etc.

[0978] Step 4:

[0979] The server preprocesses the acquired data. Specifically, it merges the number of inspections, weather information, and event information based on the date. Each data frame is required as input. The merged data integrates the number of inspections, weather information, and event information for each date.

[0980] Step 5:

[0981] The server gracefully handles missing values ​​and outliers present in the preprocessed data. The input is the merged data. Missing values ​​are imputed with zeros or the mean, and outliers are removed or corrected. The output is a clean dataset.

[0982] Step 6:

[0983] The server trains a predictive model (linear regression model) using the clean data. The inputs include features such as weather and event information, and a target variable called the number of reviews. The Scikit-learn library is used to train the model. The trained model is obtained as the output and stored on the server.

[0984] Step 7:

[0985] The server retrieves future weather and event information and preprocesses it. Similar to steps 2 and 3, it retrieves data from the API and converts it into a data frame. As input, it requires future weather and event information, and obtains preprocessed future data as output.

[0986] Step 8:

[0987] The server uses the trained linear regression model to predict future inspection volume. As input, it uses preprocessed future weather and event information. As output, it obtains a data frame of predicted inspection volume.

[0988] Step 9:

[0989] The user's device runs the emotion engine using facial expressions, voice, and text input. When the user uses the application, the emotion engine analyzes input data from the camera and microphone. The inputs required are the user's facial expression data, voice, and text input. The output is a numerical representation of the user's emotional state, which is sent to the server.

[0990] Step 10:

[0991] The server receives emotion information from the emotion engine and adjusts the prediction results. The inputs are emotion information and the predicted number of reviews. For example, if the user is feeling stressed, the server increases the predicted number of reviews. The output is the adjusted prediction results.

[0992] Step 11:

[0993] The user device presents the optimized products and services to the user. Based on the adjusted prediction results, products and promotions that the user is likely to be interested in are displayed. The adjusted prediction results are required as input, and the products and services are presented in a form that the user can see as output.

[0994] 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.

[0995] 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.

[0996] 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.

[0997] [Fourth embodiment]

[0998] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0999] 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.

[1000] 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).

[1001] 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.

[1002] 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.

[1003] 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).

[1004] 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.

[1005] 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.

[1006] 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.

[1007] 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.

[1008] 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.

[1009] 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.

[1010] 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."

[1011] This system consists of a server and a user terminal. The server acquires the necessary data from weather forecast services and event information services and performs a series of processes to train the model. The user terminal provides an interface for users to check the results.

[1012] Explanation of program processing

[1013] Data collection

[1014] 1. The server first retrieves audit data from the historical audit database, which involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame.

[1015] 2. The server sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. The weather information includes data such as temperature, humidity, and weather type for each date.

[1016] 3. The server also sends an HTTP request to an external event information service to retrieve event information for the specified region and period, including the event name, date, location, and number of participants.

[1017] Data Preprocessing

[1018] 4. The server combines the acquired inspection data, weather information, and event information to generate a unified dataset. Each data frame is merged based on date.

[1019] 5. The server will gracefully handle missing values ​​and outliers in the combined datasets, using imputation such as zeros and mean values.

[1020] Training the model

[1021] 6. The server uses the preprocessed data to train a linear regression model, which uses weather and event information as features and the number of reviews as targets.

[1022] 7. The server stores the trained model for future predictions.

[1023] Prediction of future review volume

[1024] 8. The server retrieves future weather and event information, preprocesses it, and reassembles it into a dataset ready to be fed into the model.

[1025] 9. The server uses the trained linear regression model to predict future audit volume. The prediction results are added to a data frame and can be saved or sent to a device.

[1026] Specific examples

[1027] For example, to predict next month's number of inspections, the server first retrieves data on the number of inspections over the past year. Then, it collects weather and event information for the same period along the time axis, integrates this data, and generates a training dataset. It then trains a linear regression model to predict the number of inspections based on next month's weather and event information. Finally, users can view the results on their devices.

[1028] Users can use the prediction results to, for example, efficiently allocate staff shift schedules or adjust response resources in advance. In this way, the system improves the accuracy of predictions of the number of examination cases, contributing to operational efficiency.

[1029] The processing flow will be explained below.

[1030] Step 1:

[1031] The server connects to the database of past review counts. It uses a database connection library, such as sqlite3 or sqlalchemy, to connect to the database. It issues an SQL query to retrieve past review count data from the database. Specifically, it executes a query such as SELECT FROM review_counts and stores the results in a Pandas data frame.

[1032] Step 2:

[1033] The server sends an HTTP request to the weather forecast service. Using the requests library, it sends parameters including the API key, location information, and forecast period (number of days) to the weather forecast service's API endpoint (e.g., https: / / api.weatherapi.com / v1 / forecast.json). It then obtains the JSON data returned as a response, extracts the necessary weather information (date, time, temperature, weather type, etc.), and converts it into a data frame.

[1034] Step 3:

[1035] The server also sends an HTTP request to the event information service. Using the requests library, it sends parameters including the API key, location information, and duration to the event information service's API endpoint (e.g., https: / / api.eventapi.com / v1 / events). The necessary information (event name, date and time, location, number of attendees, etc.) is extracted from the JSON data returned as a response and converted into a data frame.

[1036] Step 4:

[1037] The server combines the obtained inspection count data, weather information, and event information. It uses the Pandas merge function to combine each data frame using the date as a key. This operation generates a single combined dataset.

[1038] Step 5:

[1039] The server processes missing values ​​in the integrated dataset. It uses Pandas' fillna function to fill missing values ​​with zeros or the average of the preceding and following days. This process ensures the completeness of the dataset.

[1040] Step 6:

[1041] The server trains a linear regression model using the preprocessed data. Using Scikit-learn's LinearRegression class, it separates the features (weather information, event information) and the target (number of reviews) and trains using the model's fit method. Through this training, the model learns parameters from the given data.

[1042] Step 7:

[1043] The server saves the trained model for future predictions. The model is saved using the joblib or pickle libraries, and stored as a file on the local disk or in a database.

[1044] Step 8:

[1045] The server again obtains the weather and event information needed for future predictions. It also sends an HTTP request to obtain future data and performs preprocessing. The obtained data is again incorporated into the integrated dataset using the date as a key.

[1046] Step 9:

[1047] The server uses the trained model to predict future audit volume. It inputs the preprocessed future data into the model and gets the prediction result. The result is added as a new column in the data frame.

[1048] Step 10:

[1049] The server saves the prediction results in a database or sends them to the user's device. When saving the results in a database, an INSERT operation is performed using an appropriate SQL query. When notifying the user, the server sends the results via email or a push notification service, for example.

[1050] Step 11:

[1051] Users can check the forecast results on their devices, view the predicted number of reviews via a dedicated management screen or notification messages, and adjust staff schedules and create work plans based on the results.

[1052] Example 1

[1053] 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."

[1054] Conventional systems had low accuracy in predicting the number of screening cases, making it difficult to improve work efficiency. Furthermore, a series of processes, such as collecting data from external services, preprocessing the data, and training the prediction model, were often performed manually, resulting in a lot of work and prone to errors. Therefore, a more accurate and automated prediction system was needed.

[1055] 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.

[1056] In this invention, the server includes a means for acquiring case count data from a past database, a means for acquiring weather data from an external information service, and a means for acquiring event information from an external information service. This enables the acquired data to be automatically integrated and preprocessed. Furthermore, the preprocessed data can be used to train a prediction model, and future case counts can be predicted with high accuracy based on the prediction results. This improves business efficiency and accuracy.

[1057] The "past database" is a data storage that holds the number of past reviews and related data.

[1058] "Case data" is data that includes information regarding the number of reviews conducted within a particular period of time.

[1059] An "external information service" is an external service that the server accesses via the Internet and that provides weather-related data and forecasts.

[1060] "Weather data" is data that includes information about the weather in a particular region or time period, such as temperature, humidity, and type of weather.

[1061] An "external information service" is an external service that the server accesses via the Internet and provides information about events and other activities.

[1062] "Event information" is data that includes details related to an event, such as the event name, date, location, and number of participants.

[1063] "Preprocessing means" refers to a set of operations or techniques used to process acquired data by cleansing, filtering, integrating, etc., and converting it into a form suitable for training a predictive model.

[1064] "Means of integration" refers to a method of combining data obtained from different sources using dates as a key to create a single unified data set.

[1065] A "predictive model" is a mathematical or statistical model trained to predict future values ​​based on past data.

[1066] A "regression analysis model" is a statistical method for analyzing the relationship between independent variables and dependent variables and making predictions.

[1067] "Future volume" refers to the number of audits that will occur under specified conditions in a projected future period.

[1068] A "server" is a computer system that executes a series of processes such as data acquisition, preprocessing, model training, and prediction calculation.

[1069] The present invention is a system that obtains event count data from a past database, weather data and event information from external information services, and predicts future event counts based on this data. The system is composed of a server and a user terminal.

[1070] Server configuration and functions

[1071] 1. Data Acquisition

[1072] The server first retrieves the historical data from the database. This is done using a database connection library and an SQL query. In Python, the pandas read_sql_query function is used to retrieve the data in data frame format.

[1073] Next, the server retrieves weather data from an external information service (e.g., a weather forecast service API) by sending an HTTP request using the requests library and parsing the resulting data in JSON format.

[1074] Additionally, the server obtains event information from an external event information service (e.g., an event information API). Again, it uses the requests library to send an HTTP request and parses the returned data in JSON format.

[1075] 2. Data Preprocessing

[1076] The server converts the acquired count data, weather data, and event information into respective data frames and combines them based on the date. This combining process uses the pandas merge function. After combining, the data set is filled in with missing values ​​and outliers are processed. For the filling process, the pandas fillna method is used to fill in missing values ​​with zero or the average value.

[1077] 3. Training the model

[1078] The server trains a predictive model (regression analysis model) based on the preprocessed dataset. For training, it uses the LinearRegression class from scikit-learn, learning the model using weather data and event information as features and count data as the target. The trained model is then serialized and saved using the joblib library.

[1079] 4. Future Data Prediction

[1080] The server again retrieves and preprocesses future weather and event data. Based on this new data, it uses the trained regression analysis model to predict future event counts. The prediction results are sent to the user's device or stored locally on the server.

[1081] User terminal configuration and functions

[1082] The user terminal provides an interface for displaying the forecast results sent from the server, allowing the user to improve work efficiency and optimize resource allocation based on these forecast figures.

[1083] Specific examples

[1084] For example, when making a prediction for next month, the server retrieves the number data for the past year from the database. At the same time, it retrieves weather data for the specified period from the weather forecast service API and collects event information from the event information provision API. This data is integrated, missing values ​​are filled in, and a training dataset is generated. The server then trains a regression analysis model based on this dataset and predicts the number of events for next month by inputting future weather data and event information. Finally, users can check these prediction results on their own devices and use them in their work.

[1085] Prompt Sentence Examples

[1086] "Please predict the number of cases next month based on the number of cases over the past year, as well as weather and event information for the specified area."

[1087] As described above, the present invention provides a system that makes highly accurate predictions based on past data and external information, thereby improving business efficiency.

[1088] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1089] Step 1:

[1090] The server first connects to the historical database and retrieves the count data. Specifically, it issues an SQL query using a database connection library. For example, it executes the query "SELECT FROM Review Count WHERE Date BETWEEN '2022-01-01' AND '2022-12-31'" and retrieves the results in data frame format. The input is an SQL query, and the output is count data in data frame format.

[1091] Step 2:

[1092] The server obtains weather data from an external information service. Specifically, it uses the requests library to send an HTTP request to obtain weather information for a specified region and period. The request URL includes parameters such as the API key, region, and period. For example, it is executed as follows: "requests.get('https: / / api.weather.com / v3 / wx / forecast / daily / 5day?apiKey=YOUR_API_KEY&geocode=LAT,LON')". The input is the API request, and the output is weather data in JSON format.

[1093] Step 3:

[1094] The server obtains event information from an external information provider. Again, it uses the requests library to send an HTTP request to obtain event information for a specified region and period. For example, execute it as follows: "requests.get('https: / / api.eventservice.com / v1 / events?apiKey=YOUR_API_KEY&location=LOCATION&startDate=START_DATE&endDate=END_DATE')". The input is the API request, and the output is event information in JSON format.

[1095] Step 4:

[1096] The server integrates the acquired count data, weather data, and event information. To do this, it uses the pandas merge function to combine each data frame based on the date. For example, execute "merged_data = pd.merge(count data, weather data, on='date')". The input is the count data, weather data, and event information in data frame format, and the output is an integrated dataset.

[1097] Step 5:

[1098] The server processes missing values ​​and outliers in the merged dataset. Specifically, it uses the pandas fillna method to fill missing values. Zero or the mean value is used for this filling. For example, execute "merged_data.fillna(0, inplace=True)". The input is the merged dataset, and the output is the imputed dataset.

[1099] Step 6:

[1100] The server trains a regression analysis model using the preprocessed dataset. Using scikit-learn's LinearRegression class, the model is trained using weather data and event information as features and count data as the target. For example, this is done using "from sklearn.linear_model import LinearRegression", "model = LinearRegression()", and "model.fit(features, target)". The input is the preprocessed dataset, and the output is a trained regression analysis model.

[1101] Step 7:

[1102] The server saves the trained model. It serializes and saves the model using the joblib library. For example, run "import joblib" and "joblib.dump(model, 'model.pkl')". The input is the trained regression model, and the output is the saved model file.

[1103] Step 8:

[1104] The server retrieves future weather data and event information again and performs preprocessing. It converts them into a data frame format again and merges them based on the date. For example, execute "future_data = pd.merge(future weather data, future event data, on='date')". The input is future weather data and event information, and the output is a preprocessed future dataset.

[1105] Step 9:

[1106] The server uses the trained regression analysis model to predict future count data. Future weather data and event information are input as features into the model to obtain prediction results. For example, execute "future_predictions = model.predict(future features)". The input is a preprocessed future dataset, and the output is the predicted future count data.

[1107] Step 10:

[1108] The server adds the prediction results to a data frame and sends it to the user's device or saves it in an appropriate database. For example, it can save it as a CSV file using the pandas to_csv method or send it to the user's device through an API. The input is a data frame containing the prediction results, and the output is the saved file or the sent data.

[1109] The above are the processing steps of the program for this system.

[1110] (Application example 1)

[1111] 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."

[1112] In traditional brick-and-mortar store operations, it was difficult to accurately predict the number of customers because operations depended on weather and local event information, making it difficult to optimize staff allocation and inventory management. This also led to staff shortages and overstaffing, hindering efficient store operations. This resulted in increased costs and a decline in service quality.

[1113] 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.

[1114] In this invention, the server includes means for acquiring data from a past database, means for acquiring information from an external weather information service, means for acquiring activity information from an external activity information service, means for preprocessing the acquired data, means for training a prediction model using the preprocessed data, means for predicting the number of customers in the future, means for optimizing staffing using the predicted number of customers in the future, and user interface means for displaying the optimized staffing and inventory management. This makes it possible to accurately predict the number of customers in the future and optimally allocate staff and inventory.

[1115] A "database" is a system designed to efficiently organize and manage information and data.

[1116] "Data acquisition means" refers to the methods and processes used to collect the necessary data.

[1117] "Weather information service" means an external source that provides information about the weather and climate for a particular region or period of time.

[1118] "Activity information services" are external information sources that provide information on local events and activities.

[1119] "Preprocessing" refers to the process of converting and organizing data into a format suitable for analysis and model training.

[1120] A "predictive model" is a computational algorithm for predicting future situations based on past data and factors.

[1121] "Customer traffic forecasting method" means a method or process used to forecast future customer traffic.

[1122] "Staffing optimization measures" are methods and techniques for most efficient staffing.

[1123] A "user interface" is a part of a system that provides a means of input and output that allows a user to directly interact with the system.

[1124] "Store foot traffic" refers to the number of customers who visit a physical store within a specific period of time.

[1125] To implement this invention, a system consisting of a server and a user terminal must be constructed. The server acquires data from a past database and collects information from external weather information services and external activity information services. The user terminal acts as an interface that provides users with forecast results, optimized staffing, and inventory management information.

[1126] ---

[1127] 1. Data Collection

[1128] The server first retrieves customer visit data from a historical database by issuing SQL queries using a database connectivity library and retrieving the results as a data frame, and then retrieves weather and event information using HTTP requests from external weather information services (e.g., OpenWeatherMap API) and activity information services (e.g., Eventful API).

[1129] ---

[1130] 2. Data Preprocessing

[1131] The acquired customer visit data, weather information, and event information are integrated based on the date, and missing or outlier values ​​in the integrated dataset are filled with zeros or the average value.

[1132] ---

[1133] 3. Model training

[1134] A linear regression model is trained using the preprocessed data. This process uses Python's Scikit-learn library. Using weather and event information as features and the number of customer visits as targets, a model is constructed to predict future customer visits.

[1135] ---

[1136] 4. Future Predictions

[1137] Future weather and event information is taken, combined into a preprocessed dataset, and a trained linear regression model is used to predict future store traffic. The prediction results are added to a data frame and saved or sent to the device as a result.

[1138] ---

[1139] 5. User Interface

[1140] The user device displays predicted future customer numbers, optimized staffing, and inventory management information. Staffing plans and inventory management can be visually confirmed using a smartphone app or the robot's display.

[1141] ---

[1142] Specific examples

[1143] For example, to generate a forecast for the next month, the server first retrieves customer visitation data from the past year. Next, it collects weather and event information from the same period along the time axis, integrates this data, and generates a training dataset. It then trains a linear regression model to predict the number of customers based on the weather and event information for the next month. Finally, users can view the results on their devices and, for example, efficiently allocate staff shift schedules or adjust response resources in advance.

[1144] ---

[1145] Prompt Sentence Examples

[1146] "To predict the number of customers coming to Store A next month, we will provide the following information: number of customers visiting the store over the past year, weather information (temperature, humidity, weather type), and event information (event name, date, location, number of participants). Based on this information, please calculate the predicted number of customers coming to the store next month."

[1147] This system is expected to significantly improve the efficiency of physical store operations, as it not only predicts store visits but also enables efficient staff deployment planning and optimized inventory management.

[1148] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1149] Step 1:

[1150] The server retrieves historical customer visit data from a database. Specifically, it uses a database connection library to issue an SQL query and obtains the results as a data frame. This data includes the number of customer visits by date. The input is the database connection information and the SQL query, and the output is a data frame of the number of customer visits by date.

[1151] Step 2:

[1152] The server obtains weather information from an external weather information service. Specifically, it sends an HTTP request to the weather information API and receives weather information as a response. This weather information includes temperature, humidity, and weather type by date. The input is the API key and the request URL, and the output is a data frame of weather information by date.

[1153] Step 3:

[1154] The server obtains event information from an external activity information service. Specifically, it sends an HTTP request to the activity information API and receives event information as a response. This event information includes the event name, date, location, and number of participants. The input is the API key and request URL, and the output is a data frame of event information by date.

[1155] Step 4:

[1156] The server combines the acquired customer visit data, weather information, and event information. Specifically, it combines each data frame using the date as a key to generate a single complete dataset. The input contains each date-specific data frame, and the output contains the combined dataset.

[1157] Step 5:

[1158] The server processes missing and outliers in the merged dataset by imputing missing values ​​with zeros or the mean and filtering outliers. The input is the merged dataset and the output is the complete, processed dataset.

[1159] Step 6:

[1160] The server trains a linear regression model using the preprocessed data. Specifically, it sets weather and event information as features and the number of customer visits as targets, and trains the regression model using the Scikit-learn library. The input is a complete dataset, and the output is a trained model.

[1161] Step 7:

[1162] The server retrieves future weather and event information and preprocesses it. The future weather and event information is again retrieved from an external service via HTTP request and combined using the date as a key. The input contains future weather and event information, and the output contains a dataset for prediction.

[1163] Step 8:

[1164] The server uses the trained linear regression model to predict future customer visits. It inputs preprocessed future weather and event information into the model to generate a prediction result. The input includes a dataset for prediction, and the output includes a prediction result of future customer visits.

[1165] Step 9:

[1166] The terminal receives the forecast data sent from the server and displays it on the user interface. Specifically, it has the function of visually displaying the predicted number of future customer visits, optimal staffing, and inventory management information. The input includes the forecast result data, and the output includes the information displayed on the user interface.

[1167] Step 10:

[1168] Users make decisions based on the forecast data through their devices. For example, they can adjust staff shift schedules or plan inventory replenishment based on the predicted number of customers. The input includes the information displayed on the device, and the output includes the user's decisions and actions.

[1169] 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.

[1170] This system consists of a server and a user terminal, and is further combined with an emotion engine that recognizes the user's emotions. The server obtains the necessary data from weather forecast services and event information services and performs a series of processes to train the model. The emotion engine also recognizes the user's emotions and adjusts the prediction results based on them. The user terminal provides an interface for the user to check the results.

[1171] Explanation of program processing

[1172] Data collection

[1173] 1. The server first retrieves audit data from the historical audit database, which involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame.

[1174] 2. The server sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. The weather information includes data such as temperature, humidity, and weather type for each date.

[1175] 3. The server also sends an HTTP request to an external event information service to retrieve event information for the specified region and period, including the event name, date, location, and number of participants.

[1176] Data Preprocessing

[1177] 4. The server combines the acquired inspection data, weather information, and event information to generate a unified dataset. Each data frame is merged based on date.

[1178] 5. The server will gracefully handle missing values ​​and outliers in the combined datasets, using imputation such as zeros and mean values.

[1179] Training the model

[1180] 6. The server uses the preprocessed data to train a linear regression model, which uses weather and event information as features and the number of reviews as targets.

[1181] 7. The server stores the trained model for future predictions.

[1182] Prediction of future review volume

[1183] 8. The server retrieves future weather and event information, preprocesses it, and reassembles it into a dataset ready to be fed into the model.

[1184] 9. The server uses the trained linear regression model to predict future audit volume. The predictions are added as new columns to the data frame.

[1185] Emotion recognition and predictive outcome adjustment

[1186] 10. The device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data and recognizes the user's emotions.

[1187] 11. The server receives emotion information from the emotion engine and adjusts the prediction results. For example, if the user is feeling stressed, the server may increase the predicted number of reviews.

[1188] Specific examples

[1189] For example, the server collects data on the number of inspections over the past year, while also collecting weather forecasts and event information for the coming month. This data is used to train a linear regression model to predict the number of inspections for the next month. The prediction results are sent to the user's device, where the user can check them. At that time, the emotion engine analyzes the user's facial expression data and voice instructions in real time and provides feedback on the user's emotional state.

[1190] For example, if the user is tired, the system will display a warning on the screen showing the prediction results and suggest increasing the number of reviews, while if the user is relaxed, the system will display standard prediction results.

[1191] In this way, a system incorporating an emotion engine can provide flexible predictions of the number of cases reviewed that take into account the user's emotional state. This functionality not only improves operational efficiency but also the user experience.

[1192] The processing flow will be explained below.

[1193] Step 1:

[1194] The server connects to the database of past review counts. It uses a database connection library, such as sqlite3 or sqlalchemy, to connect to the database. It issues an SQL query to retrieve past review count data from the database. Specifically, it executes a query such as SELECT FROM review_counts and stores the results in a Pandas data frame.

[1195] Step 2:

[1196] The server sends an HTTP request to an external weather forecast service. Using the requests library, it sends parameters including the API key, location information, and forecast period (number of days) to the weather forecast service's API endpoint (e.g., https: / / api.weatherapi.com / v1 / forecast.json). It then obtains the JSON data returned as a response, extracts the necessary weather information (date, time, temperature, weather type, etc.), and converts it into a data frame.

[1197] Step 3:

[1198] The server also sends an HTTP request to the event information service. Using the requests library, it sends parameters including the API key, location information, and duration to the event information service's API endpoint (e.g., https: / / api.eventapi.com / v1 / events). The necessary information (event name, date and time, location, number of attendees, etc.) is extracted from the JSON data returned as a response and converted into a data frame.

[1199] Step 4:

[1200] The server combines the obtained inspection count data, weather information, and event information. It uses the Pandas merge function to combine each data frame using the date as a key. This operation generates a single combined dataset.

[1201] Step 5:

[1202] The server processes missing values ​​in the integrated dataset. It uses Pandas' fillna function to fill missing values ​​with zeros or the average of the preceding and following days. This process ensures the completeness of the dataset.

[1203] Step 6:

[1204] The server trains a linear regression model using the preprocessed data. Using Scikit-learn's LinearRegression class, it separates the features (weather information, event information) and the target (number of reviews) and trains using the model's fit method. Through this training, the model learns parameters from the given data.

[1205] Step 7:

[1206] The server saves the trained model for future predictions. The model is saved using the joblib or pickle libraries, and stored as a file on the local disk or in a database.

[1207] Step 8:

[1208] The server again obtains the weather and event information needed for future predictions. It also sends an HTTP request to obtain future data and performs preprocessing. The obtained data is again incorporated into the integrated dataset using the date as a key.

[1209] Step 9:

[1210] The server uses the trained model to predict future audit volume. It inputs the preprocessed future data into the model and gets the prediction result. The result is added as a new column in the data frame.

[1211] Step 10:

[1212] The server saves the prediction results in a database or sends them to the user's device. When saving the results in a database, an INSERT operation is performed using an appropriate SQL query. When notifying the user, the server sends the results via email or a push notification service, for example.

[1213] Step 11:

[1214] The device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data in real time and recognizes the user's emotions. For example, it uses machine learning libraries and APIs to perform facial recognition and voice analysis.

[1215] Step 12:

[1216] The server receives emotional information from the emotion engine and adjusts the prediction results. For example, if the server detects that the user is feeling stressed, it may increase the predicted number of reviews. This adjustment is achieved by using weights and correction values ​​to reflect the emotional information.

[1217] Step 13:

[1218] Users can check the prediction results on their devices. They can view the predicted number of reviews via a dedicated management screen or notification messages, and use the results to adjust staff schedules and make work plans. They can also provide feedback on the prediction results, contributing to improving the system's prediction accuracy and user experience.

[1219] Example 2

[1220] 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."

[1221] Conventional systems for predicting the number of review cases make predictions based on past data and external information, but they have the problem of being unable to make flexible predictions that take into account the emotional state of the user. In particular, ignoring the impact of user emotions on prediction results can lead to problems such as reduced operational efficiency and a lower user experience.

[1222] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring information on the number of processed cases from a past information acquisition device, means for acquiring weather information from an external weather information acquisition device, means for acquiring event information from an external event information acquisition device, means for preprocessing the acquired information, means for training a prediction model using the preprocessed information, means for predicting the number of processed cases in the future, means for recognizing emotions using an emotion recognition device, and means for adjusting the prediction result based on the recognized emotion information. This makes it possible to reflect the user's emotions in the prediction result, which is expected to improve the user experience and business efficiency.

[1223] The "past information acquisition device" is a means for acquiring necessary information from a database or data storage that stores past information on the number of processed cases.

[1224] The "external weather information acquisition device" is a means for acquiring weather information for a specified area or period from a weather information providing service on the Internet.

[1225] An "external event information acquisition device" is a means for acquiring event information for a specified area or period from an event information providing service on the Internet.

[1226] "Preprocessing" is the operation of integrating the acquired information, appropriately handling missing values ​​and outliers, and converting it into a format suitable for training a predictive model.

[1227] A "predictive model" is a mathematical or statistical model used to predict future transaction volumes based on past and external information.

[1228] An "emotion recognition device" is a means for analyzing a user's facial expressions, voice, character input, etc. to recognize the user's emotional state.

[1229] "Emotion information" is data relating to the user's emotional state obtained by an emotion recognition device.

[1230] "Adjusting the predicted result" is an operation for correcting or amending the predicted number of transactions in accordance with the emotion based on the recognized emotion information.

[1231] The system of the present invention includes a past information acquisition device, an external weather information acquisition device, an external event information acquisition device, a preprocessing device, a prediction model device, an emotion recognition device, and a prediction result adjustment device. Each device and its specific operation are described below.

[1232] First, the server uses the past information acquisition device to acquire the past processing count information from the database. To do this, for example, it uses the Python pandas library to issue an SQL query and acquire the data as a data frame.

[1233] Next, the server uses an external weather information acquisition device to send an HTTP request to the weather information service to obtain weather information for the specified area and period. This operation uses the requests library to obtain information such as temperature, humidity, and weather type.

[1234] The server then uses an external event information acquisition device to send an HTTP request to the event information service to obtain event information for the specified region and period. Again, the requests library is used to collect data such as the event name, date, location, and number of participants.

[1235] The acquired information is preprocessed by the server's preprocessing unit. Specifically, the data frames are merged based on the date using the pandas merge function to integrate the dataset. Missing and outlier values ​​are imputed using the pandas fillna method.

[1236] Next, based on the preprocessed data, the server trains a predictive model using the predictive modeler. Here, a linear regression model is trained using the LinearRegression class of the scikit-learn library. The trained model is saved using the joblib library for future predictions.

[1237] The server retrieves future weather and event information again, preprocesses it, and then uses the trained predictive model to predict future transaction volumes. The prediction results are added as new columns to a pandas data frame.

[1238] Furthermore, the user device runs an emotion recognition device using the user's facial expressions, voice, and text input. Here, the OpenCV library and speech recognition library are used to acquire data in real time and recognize the user's emotions. The recognized emotion information is reflected in the prediction results by the prediction result adjustment device on the server. For example, if the user is feeling stressed, the predicted number of transactions is adjusted to increase.

[1239] Specific examples

[1240] For example, the server retrieves transaction volume information from the database for the previous year, then collects weather and event information for the coming month. Based on this information, a linear regression model is trained using the LinearRegression class in scikit-learn to predict transaction volume for the next month. The results are then sent to the user's device, where the user can view the predictions.

[1241] If a user is tired while using the device, the device will perform facial recognition and voice analysis, and the emotion recognition device will detect the emotion "tired." This information will be sent to the server, and the prediction results will be adjusted. For example, a warning based on the prediction results may be displayed, suggesting an increase in the number of reviews.

[1242] Prompt Sentence Examples

[1243] "Please build a system that predicts the number of cases to be processed next month based on weather and event information."

[1244] "Please add a feature to adjust the predicted transaction volume results based on the user's emotional state."

[1245] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1246] Step 1:

[1247] The server uses the past information acquisition device to obtain information on the number of past processed items from the database. Database connection information and an SQL query are used as input to obtain information on the number of past processed items from the database. Specifically, an SQL query is issued using the Python pandas library, and the results are obtained as a data frame. The output is a data frame containing information on the number of past processed items.

[1248] Step 2:

[1249] The server uses an external weather information acquisition device to send an HTTP request to the weather information service to obtain weather information for the specified region and period. The region, period, and API key are used as input. The API is accessed through the requests library to obtain weather information. The output is data including information such as temperature, humidity, and weather type for each date.

[1250] Step 3:

[1251] The server uses an external event information acquisition device to send an HTTP request to the event information service to obtain event information for the specified region and period. This also uses the region and period information and the API key as input. The requests library is used to access the API and obtain the event information. The output includes information such as the event name, date, location, and number of participants.

[1252] Step 4:

[1253] The server's preprocessing unit combines the acquired transaction count information, weather information, and event information to generate an integrated dataset. It uses a data frame of past transaction count information, weather information data, and event information data as input. It uses the pandas merge function to merge these data frames based on the date. The output is an integrated dataset.

[1254] Step 5:

[1255] The server uses a preprocessor to appropriately handle missing values ​​and outliers in the combined datasets. The input is the combined dataset, and missing and outliers are filled with zeros or the mean value using the pandas fillna method. The output is a dataset with missing and outliers processed.

[1256] Step 6:

[1257] The server trains a predictive model using the preprocessed data. The input is the preprocessed dataset. Using the LinearRegression class from the scikit-learn library, it sets weather information and event information as features and the number of items to be processed as the target. Specifically, it uses the fit method of the model to train the model. The output is the trained predictive model.

[1258] Step 7:

[1259] The server saves the trained predictive model. The input is the trained predictive model. It uses the joblib library to save the model to the file system. The output is the saved model file.

[1260] Step 8:

[1261] The server again retrieves and preprocesses future weather and event information. The input is a future weather information API request and an event information API request. Again, it uses the requests library to retrieve the information and perform preprocessing operations. The output is a dataset containing preprocessed future weather and event information.

[1262] Step 9:

[1263] The server uses the trained prediction model to predict future transaction volumes. The input is preprocessed future weather and event information. The predict method of the prediction model is used to predict future transaction volumes, and the prediction results are added as new columns in a pandas data frame. The output is a dataset with predicted future transaction volumes.

[1264] Step 10:

[1265] The user device runs an emotion recognition device using facial expressions, voice, and text input. The inputs are the user's facial expression data, voice data, and text input data. Using the OpenCV library and voice recognition library, emotions are recognized in real time. The output is the recognized emotion information.

[1266] Step 11:

[1267] The server receives emotion information from the emotion recognition device and adjusts the prediction results. The inputs are the predicted number of transactions and emotion information. For example, if the user is feeling stressed, the server makes adjustments such as increasing the prediction value. The emotion information is reflected in the prediction results and the final prediction results are output.

[1268] (Application example 2)

[1269] 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."

[1270] Conventional prediction systems often make predictions without considering the user's emotions, which often results in insufficient consideration of the user experience. Furthermore, there is a need for systems that can recognize the user's emotional state and respond flexibly based on that. However, conventional technologies lack prediction systems that incorporate emotion recognition, and there is also a lack of means to effectively utilize the prediction results.

[1271] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1272] In this invention, the server includes means for acquiring data on the number of audits from a past database, means for acquiring weather information from an external weather forecast service, means for acquiring event information from an external event information service, means for preprocessing the acquired data, means for training a prediction model using the preprocessed data, means for predicting the number of audits in the future, means for recognizing user emotions and adjusting the prediction results, and means for presenting individually optimized products and services. This enables flexible responses that take user emotions into consideration, makes it possible to more effectively utilize the prediction results, and provides a high-quality user experience.

[1273] "Data on number of examinations" refers to data that includes quantities and statistical information related to past examinations.

[1274] "Weather information" refers to data such as temperature, humidity, and weather type obtained from an external weather forecast service.

[1275] "Event information" refers to data such as the event name, date, venue, and number of participants obtained from an external event information service.

[1276] "Preprocessing" refers to the process of integrating acquired data, treating missing values, and treating outliers.

[1277] A "predictive model" is a mathematical model that is trained based on preprocessed data and is used to predict future numbers of reviews, etc.

[1278] "Emotion recognition" is a technology that analyzes a user's emotions from their facial expressions, voice, text input, etc.

[1279] "Presenting individually optimized products and services" means suggesting optimal products and services in real time based on the user's emotional state and acquired data.

[1280] This invention is a system that is composed of a server and a user terminal, and further combines an emotion engine that recognizes the user's emotions. The server is connected to the Internet and performs the following specific functions.

[1281] First, the server retrieves the historical audit count data from the database. This involves issuing an SQL query using a database connectivity library and retrieving the results as a data frame. The server then sends an HTTP request to an external weather forecast service to retrieve weather information for the specified region and time period. It also sends an HTTP request to an external event information service to retrieve event information for the specified region and time period. This data is then consolidated on the server, with missing values ​​and outliers being handled appropriately.

[1282] The server then uses the preprocessed data to train a predictive model, specifically a linear regression model, with weather and event information as features and the number of audits as targets. The trained model is stored on the server for future predictions.

[1283] Meanwhile, the user device runs an emotion engine using the user's facial expressions, voice, and text input. The emotion engine analyzes this data and recognizes the user's emotions. When the user accesses the virtual store, the emotion engine performs real-time analysis and provides feedback to the server to adjust the prediction results.

[1284] For example, consider a scenario in which a user is shopping online. The user's device uses a smartphone or head-mounted display (HMD) to display products that the user may be interested in. Based on the user's purchasing history, weather information, event information, and emotional data, the server suggests optimal products and promotions in real time. If the user is feeling stressed, the number and variety of suggested products can be adjusted, improving the user experience.

[1285] Possible hardware includes a database server, an API server, and user devices (smartphones, smart glasses, HMDs). Software includes Python, Pandas, Scikit-learn, and virtual libraries for emotion recognition (e.g., SentimentEngine). API communication uses libraries such as the requests library.

[1286] As a concrete example, the server acquires data on the number of inspections over the past year and collects weather forecasts and event information for the coming month. This data is used to train a linear regression model to predict the number of inspections for the next month. The prediction results are sent to the user's device, where the user can check them. At that time, the emotion engine analyzes the user's facial expression data and voice instructions in real time and provides feedback on the user's emotional state.

[1287] Prompt Sentence Examples

[1288] Implement an example of a product recommendation system that takes user emotions into account. Recognize emotions in real time based on past purchase history, weather information, and event information, and suggest optimized products and services.

[1289] Required parts:

[1290] 1. Data Collection and Preprocessing

[1291] 2. Training a model based on the data

[1292] 3. Emotion Recognition Engine Integration

[1293] 4. Real-time optimized recommendations

[1294] Please implement a concrete example in Python.

[1295] By implementing a system based on this prompt, it becomes possible to make suggestions that take into account the user's emotions in real time, providing a high-quality user experience.

[1296] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1297] Step 1:

[1298] The server retrieves historical audit count data from the database. For input, it uses a database connection library and issues an SQL query. The audit count data retrieved from the database is output as a data frame. This data includes the number of audits by date.

[1299] Step 2:

[1300] The server sends an HTTP request to an external weather forecast service to obtain weather information for a specified region and period. The API endpoint and request parameters are required as input. The obtained weather information is output to the server in JSON format and converted into a data frame using a data analysis library. The weather information includes data such as temperature, humidity, and weather type for each date.

[1301] Step 3:

[1302] The server sends an HTTP request to an external event information service to obtain event information for a specified region and period. The API endpoint and request parameters are required as input. The obtained event information is output to the server in JSON format and converted into a data frame using a data analysis library. The event information includes the event name, date, location, number of attendees, etc.

[1303] Step 4:

[1304] The server preprocesses the acquired data. Specifically, it merges the number of inspections, weather information, and event information based on the date. Each data frame is required as input. The merged data integrates the number of inspections, weather information, and event information for each date.

[1305] Step 5:

[1306] The server gracefully handles missing values ​​and outliers present in the preprocessed data. The input is the merged data. Missing values ​​are imputed with zeros or the mean, and outliers are removed or corrected. The output is a clean dataset.

[1307] Step 6:

[1308] The server trains a predictive model (linear regression model) using the clean data. The inputs include features such as weather and event information, and a target variable called the number of reviews. The Scikit-learn library is used to train the model. The trained model is obtained as the output and stored on the server.

[1309] Step 7:

[1310] The server retrieves future weather and event information and preprocesses it. Similar to steps 2 and 3, it retrieves data from the API and converts it into a data frame. As input, it requires future weather and event information, and obtains preprocessed future data as output.

[1311] Step 8:

[1312] The server uses the trained linear regression model to predict future inspection volume. As input, it uses preprocessed future weather and event information. As output, it obtains a data frame of predicted inspection volume.

[1313] Step 9:

[1314] The user's device runs the emotion engine using facial expressions, voice, and text input. When the user uses the application, the emotion engine analyzes input data from the camera and microphone. The inputs required are the user's facial expression data, voice, and text input. The output is a numerical representation of the user's emotional state, which is sent to the server.

[1315] Step 10:

[1316] The server receives emotion information from the emotion engine and adjusts the prediction results. The inputs are emotion information and the predicted number of reviews. For example, if the user is feeling stressed, the server increases the predicted number of reviews. The output is the adjusted prediction results.

[1317] Step 11:

[1318] The user device presents the optimized products and services to the user. Based on the adjusted prediction results, products and promotions that the user is likely to be interested in are displayed. The adjusted prediction results are required as input, and the products and services are presented in a form that the user can see as output.

[1319] 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.

[1320] 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.

[1321] 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 robot 414.

[1322] 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.

[1323] FIG. 9 is a diagram illustrating 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 actions 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.

[1324] 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.

[1325] 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).

[1326] 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.

[1327] 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."

[1328] 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.

[1329] 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).

[1330] 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.

[1331] 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.

[1332] 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.

[1333] 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.

[1334] 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.

[1335] 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.

[1336] 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.

[1337] 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.

[1338] 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.

[1339] 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.

[1340] The following is further disclosed regarding the above embodiment.

[1341] (Claim 1)

[1342] A means of obtaining data on the number of examinations from past databases, and

[1343] A means for obtaining weather information from an external weather forecast service;

[1344] A means for acquiring event information from an external event information service;

[1345] means for pre-processing the acquired data;

[1346] means for training a predictive model using the preprocessed data;

[1347] A means of predicting future review volume;

[1348] A system including:

[1349] (Claim 2)

[1350] 10. The system of claim 1,

[1351] The system wherein the pre-processing means comprises means for consolidating the acquired data based on date.

[1352] (Claim 3)

[1353] 10. The system of claim 1,

[1354] The system wherein the predictive model is a linear regression model.

[1355] "Example 1"

[1356] (Claim 1)

[1357] A means of obtaining count data from the historical database;

[1358] a means for obtaining weather data from an external information service;

[1359] A means for acquiring event information from an external information providing service;

[1360] means for pre-processing the acquired data;

[1361] means for training a predictive model using the preprocessed data;

[1362] A means of predicting future numbers,

[1363] A system including:

[1364] (Claim 2)

[1365] 10. The system of claim 1, wherein the preprocessing means comprises means for consolidating the acquired data based on date.

[1366] (Claim 3)

[1367] 2. The system of claim 1, wherein the predictive model is a regression analysis model.

[1368] "Application Example 1"

[1369] (Claim 1)

[1370] a means of retrieving data from the historical database;

[1371] A means for obtaining information from an external weather information service;

[1372] A means for acquiring activity information from an external activity information service;

[1373] means for pre-processing the acquired data;

[1374] means for training a predictive model using the preprocessed data;

[1375] A means of predicting future customer numbers,

[1376] A means for optimizing staffing using the predicted future customer numbers;

[1377] a user interface means for displaying optimized staffing and inventory management;

[1378] A system including:

[1379] (Claim 2)

[1380] 10. The system of claim 1, further comprising means for aggregating the acquired data based on date.

[1381] (Claim 3)

[1382] 2. The system of claim 1, wherein the predictive model is a regression model.

[1383] "Example 2: Combining Emotion Engines"

[1384] (Claim 1)

[1385] A means for acquiring information on the number of processed cases from a past information acquisition device;

[1386] means for acquiring weather information from an external weather information acquisition device;

[1387] means for acquiring event information from an external event information acquisition device;

[1388] means for pre-processing the acquired information;

[1389] means for training a predictive model using the preprocessed information;

[1390] A means of predicting future transaction volumes;

[1391] means for recognizing emotions using an emotion recognition device;

[1392] a means for adjusting the prediction result based on the recognized emotion information;

[1393] A system including:

[1394] (Claim 2)

[1395] 2. The system of claim 1, wherein the preprocessing means comprises means for consolidating the acquired information based on date.

[1396] (Claim 3)

[1397] 2. The system of claim 1, wherein the predictive model is a statistical predictive model.

[1398] "Application example 2 when combining emotion engines"

[1399] (Claim 1)

[1400] A means of obtaining data on the number of examinations from past databases, and

[1401] A means for obtaining weather information from an external weather forecast service;

[1402] A means for acquiring event information from an external event information service;

[1403] means for pre-processing the acquired data;

[1404] means for training a predictive model using the preprocessed data;

[1405] A means of predicting future review volume;

[1406] means for recognizing a user's emotions and adjusting the prediction results;

[1407] A means of presenting individually optimized products and services,

[1408] A system including:

[1409] (Claim 2)

[1410] 10. The system of claim 1, further comprising means for aggregating the acquired data based on date.

[1411] (Claim 3)

[1412] 10. The system of claim 1, further comprising means for using the preprocessed data to train a model for making predictions based on a user's purchasing history, weather information, event information, and emotion data. [Explanation of symbols]

[1413] 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. A means of obtaining data on the number of examinations from past databases, and A means for obtaining weather information from an external weather forecast service; A means for acquiring event information from an external event information service; means for pre-processing the acquired data; means for training a predictive model using the preprocessed data; A means of predicting future review volume; A system including:

2. 10. The system of claim 1, The system wherein the pre-processing means comprises means for consolidating the acquired data based on date.

3. 10. The system of claim 1, The system wherein the predictive model is a linear regression model.

Images