system

Abstract

Provide a system. 【Solution means】 Means for collecting pet behavior data using sensors and cameras, Means for temporarily storing the collected behavior data, Means for sending the stored behavior data to a server, Means for analyzing the sent behavior data and applying a machine learning model for detecting abnormalities, Means for notifying the terminal of the detected abnormality information, Means for displaying the notified abnormality information and providing appropriate countermeasures to the user, Means for collecting user feedback and sending it to the server, Means for retraining the machine learning model using the sent feedback, A system including the above.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] Pets cannot understand human language, so there is a problem that it is difficult to detect the health status and discomfort of pets at an early stage. In particular, if the owner is slow to notice the abnormal behavior of the pet, the risk of deterioration of the pet's health increases. To solve this problem, a system that automatically monitors the behavior of pets and detects abnormalities at an early stage is required.

Means for Solving the Problems

[0005] The present invention solves the above problems by providing a system that includes means for collecting pet behavior data using sensors and cameras, means for temporarily storing the collected behavior data, means for transmitting the stored behavior data to a server, means for analyzing the transmitted behavior data and applying a machine learning model to detect anomalies, means for notifying a terminal of the detected anomaly, means for displaying the notified anomaly information and providing the user with appropriate countermeasures, means for collecting user feedback and transmitting it to a server, and means for retraining the machine learning model using the transmitted feedback.

[0006] A "sensor" is a device that detects physical states or changes and outputs that information as an electrical signal.

[0007] A "camera" is a device that records light as digital data and captures images.

[0008] "Behavioral data" refers to data that records a series of activities of a pet over time, such as movement, eating, sleeping, and excretion.

[0009] "Temporary storage" means storing data only for a short period necessary for processing or transfer, rather than storing it long-term.

[0010] A "server" is a computer system that receives and stores data sent from multiple clients, and performs data analysis and processing.

[0011] "Transmission" refers to the movement of data or information from one device to another.

[0012] A "machine learning model" is an algorithm used to analyze and predict specific patterns and trends based on large amounts of data.

[0013] "Analysis" is the process of thoroughly investigating and evaluating data to extract specific information or patterns.

[0014] "Abnormality" refers to a state or operation that is different from normal or outside the expected range.

[0015] "Terminal" refers to a device for communicating with a server to send, receive, and process data.

[0016] "Notification" means to inform a user of specific information or a state.

[0017] "User" refers to a person who uses this system, specifically the owner of a pet.

[0018] "Feedback" refers to information, opinions, and observation results provided by a user.

[0019] "Retraining" is a process for improving the performance of a machine learning model using new data and feedback.

[0020] "Countermeasure" refers to a proposal or means for dealing with a specific problem.

Brief Description of Drawings

[0021] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7]It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Mode for Carrying Out the Invention

[0022] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0023] First, the language used in the following description will be explained.

[0024] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0025] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0026] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0027] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0028] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0029] [First Embodiment]

[0030] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0031] As shown in Figure 1, the 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.

[0032] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0033] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0034] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0035] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0036] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0037] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0038] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0039] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0040] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0041] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0042] This invention is a system designed to detect the health status and illnesses of pets at an early stage. This system collects behavioral data from pets using multiple sensors and cameras, and analyzes this data on a server to detect abnormalities. The specific configuration and operation of this system are described in detail below.

[0043] Data collection

[0044] The device collects behavioral data in real time via sensors and cameras attached to the pet. For example, it tracks the pet's movement patterns with a GPS sensor and records its activity level with an accelerometer. It captures the pet's behavior as video using a camera, and records the pet's movements as video data. This data is temporarily stored in a buffer within the device.

[0045] Data transmission

[0046] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security. The data is time-stamped to maintain its order precisely.

[0047] Data analysis and anomaly detection

[0048] The server analyzes the received data. First, it converts the data into a format that can be analyzed, and then cleans it by checking for outliers and missing values. Next, it extracts features to compare with past behavioral data. Specific features include exercise levels, meal frequency, and toilet usage frequency. Based on these features, a machine learning model is used to detect anomalies. When an abnormal behavioral pattern is detected, information indicating that the behavior is abnormal is generated.

[0049] Anomaly notification and countermeasures provided

[0050] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the detected abnormal behavior. The terminal displays this information to the user (pet owner) through a user interface and provides appropriate countermeasures. For example, if an anomaly is detected, an alert such as "Your pet is behaving unusually. Please check on them" is displayed to the user.

[0051] User behavior and feedback

[0052] Based on the notified abnormal information, the user checks the pet's current condition and takes appropriate action as needed. The user observes the pet's condition and takes action, such as taking it to a veterinarian. Afterwards, they provide feedback through the application regarding the actual observation results and the actions taken.

[0053] Gathering feedback and retraining the model

[0054] The device sends feedback information provided by the user to the server. The server then uses this feedback information to retrain the anomaly detection model, improving the system's accuracy. This allows for more accurate monitoring of the pet's health.

[0055] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. The device then sends a notification to the user stating, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian." Based on this information, the user can check their pet's paws and, if there is a problem, take it to a veterinarian, enabling early detection and treatment.

[0056] The following describes the processing flow.

[0057] Step 1:

[0058] The device collects behavioral data in real time using sensors and cameras attached to the pet. Specifically, it tracks movement patterns with a GPS sensor and records activity levels with an accelerometer. At the same time, it captures behavior as video with a camera and detects specific actions (e.g., walking, sitting, licking).

[0059] Step 2:

[0060] The device temporarily stores the collected data in a buffer. The buffer is time-stamped to accurately preserve the order of the data. For example, movement data, activity level data, and video data are stored in chronological order.

[0061] Step 3:

[0062] The terminal sends data from a buffer to the server at regular intervals. This transmission process uses secure protocols such as TLS to ensure data consistency and security. For example, data batches are sent to the server every minute.

[0063] Step 4:

[0064] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and normalizes the sensor's numerical data. For example, it converts the video data into JPEG frames and scales the acceleration data within a specified range.

[0065] Step 5:

[0066] The server performs data cleaning. It checks for outliers (e.g., obviously incorrect numbers or corrupted data) and missing values, and removes or fills in the invalid data. For example, it might fill in abnormal spikes in activity level data or missing consecutive frames.

[0067] Step 6:

[0068] The server extracts features to compare with past pet behavior data stored in the database. For example, it calculates features such as daily exercise, number of meals, and frequency of litter box use.

[0069] Step 7:

[0070] The server uses a machine learning model to analyze features and detect anomalies. Specifically, it compares abnormal behavior (e.g., continuous licking of paws, sudden decrease in activity level) with normal behavior patterns to detect abnormal behavior. The model is trained based on past data and feedback.

[0071] Step 8:

[0072] The server notifies the terminal of the detected anomaly. The anomaly information includes details such as the type of abnormal behavior detected, the date and time, and the frequency.

[0073] Step 9:

[0074] The device notifies the user of any abnormal information it receives via the user interface. For example, a smartphone application might display a pop-up notification with a message such as, "Your pet is behaving unusually. Please check on it."

[0075] Step 10:

[0076] The user checks the notification, observes their pet's condition, and takes necessary action. For example, they might check their pet's paws and consult a veterinarian if any abnormalities are found.

[0077] Step 11:

[0078] Users provide feedback through the application, including observations and actions taken. Specifically, they input text information about their pet's current condition, as well as veterinary diagnoses.

[0079] Step 12:

[0080] The device sends user feedback information to the server. The transmitted feedback includes specific observations and details of treatment.

[0081] Step 13:

[0082] The server analyzes the received feedback information and retrains the machine learning model based on the new insights. This improves the system's anomaly detection accuracy and provides more precise diagnoses.

[0083] (Example 1)

[0084] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0085] In modern society, it is extremely important to monitor pets' health in real time and detect abnormalities early. However, with conventional systems, pet owners often fail to notice their pets' illnesses until it is too late, making early treatment difficult. Furthermore, there is a lack of efficient systems for collecting and analyzing pet behavioral data and developing appropriate countermeasures based on that data.

[0086] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0087] In this invention, the server includes means for collecting pet behavior data using a detection device, means for temporarily storing the collected behavior data, means for transmitting the stored behavior data to an information processing device, means for analyzing the transmitted behavior data and applying a machine learning algorithm to detect anomalies, means for notifying a display terminal of the detected anomaly information, means for displaying the notified anomaly information and providing the user with appropriate countermeasures, means for collecting user feedback and transmitting it to the information processing device, and means for retraining the machine learning algorithm using the transmitted feedback. This enables real-time monitoring of the pet's health status, early detection of anomalies, and prompt and appropriate responses.

[0088] A "detection device" is a device that includes sensors and cameras used to collect pet behavioral data.

[0089] A "location detection sensor" is a sensor used to obtain a pet's location information, and specifically includes GPS sensors.

[0090] A "activity detection sensor" is a sensor used to measure a pet's activity level, and specifically includes accelerometers and similar sensors.

[0091] "Behavioral data" refers to data including the pet's location, movements, and activity level.

[0092] "Means of temporary storage" refers to memory or storage for temporarily storing collected data.

[0093] An "information processing device" is a computing device that analyzes received data and outputs the results.

[0094] A "machine learning algorithm" is a mathematical model used to detect anomalies based on past data.

[0095] A "display terminal" is a device used to notify users of abnormal information transmitted from a server. Specifically, this includes smartphones and tablets.

[0096] "Users" refer to pet owners or caretakers.

[0097] "Feedback" refers to information about what users observed and the actions they took.

[0098] This invention relates to a system for early detection of the health status or illness of pets. This system uses multiple hardware and software components to collect and analyze pet behavioral data and detect abnormalities. Specifically, it uses a combination of sensors and cameras as detection devices, an information processing device for analysis and notification, and a terminal for collecting feedback.

[0099] The device collects real-time behavioral data of the pet using location detection sensors (e.g., GPS sensors) and activity detection sensors (e.g., accelerometers) attached to the pet. This data is stored in the device's temporary storage. The collected data is analyzed and notifications are sent according to the following processing flow.

[0100] Hardware and software usage

[0101] 1. Detection device:

[0102] Location sensors collect information about a pet's current location. For example, a GPS sensor tracks the pet's movement patterns.

[0103] The activity detection sensor records the pet's activity level and uses an accelerometer to measure how active the pet is.

[0104] 2. Information processing device:

[0105] Data transmission: The terminal sends the collected data to the server (information processing device) at regular intervals (e.g., every 5 minutes). This process is carried out using a secure protocol (e.g., TLS).

[0106] Data Analysis and Anomaly Detection: The server analyzes the received data and detects anomalies. First, it converts the data into a parseable format (e.g., CSV or JSON). Next, it cleans the data and extracts features (such as exercise levels, meal frequency, and toilet usage frequency). Based on these features, it applies machine learning algorithms to detect abnormal behavior.

[0107] 3. Notice:

[0108] Anomaly Notification: The server notifies the terminal of any detected anomalies. The terminal displays this information to the user through a user interface (such as a smartphone app). For example, an alert might be displayed stating, "Your pet is behaving unusually. Please check on it."

[0109] 4. Gathering feedback:

[0110] User Behavior and Feedback: Users check the status of their pets based on the notified abnormal information and take appropriate measures as needed. Users provide feedback on their observations and actions taken through the application.

[0111] Model retraining: The terminal sends the provided feedback information to the server, which then retrains the anomaly detection model based on this feedback. This improves the system's accuracy and enables more precise anomaly detection in the future.

[0112] Specific example

[0113] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws, and this video data is sent from the device to a server. The server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. As a result, the device sends a notification to the user stating, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian." Based on this information, the user can check their pet's paws and, if there is a problem, take it to a veterinarian, enabling early detection and treatment.

[0114] An example of a prompt statement is: "Use a system that collects pet behavior data using cameras and sensors and analyzes it on a server to monitor the pet's health in real time and detect abnormalities."

[0115] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0116] Step 1:

[0117] Data collection

[0118] The device collects behavioral data in real time using location sensors (e.g., GPS sensors) and activity sensors (e.g., accelerometers) attached to the pet. Inputs include signals from the sensors (location information, acceleration data). These sensor signals are converted into digital data and stored in a temporary buffer memory within the device. Outputs consist of a set of location information and activity data.

[0119] Specific actions:

[0120] The GPS sensor collects location information every second.

[0121] The accelerometer records the pet's movements every 0.1 seconds.

[0122] The collected data is temporarily stored in a buffer.

[0123] Step 2:

[0124] Data transmission

[0125] The terminal sends data stored in a buffer to the server at regular intervals (e.g., every 5 minutes). The input is location and momentum data stored in the buffer, and the output is data packets sent to the server. Secure protocols such as TLS are used for transmission, and the order of the data is maintained by timestamping each data packet.

[0126] Specific actions:

[0127] The data stored in the buffer is read and sent to the server using the TLS protocol.

[0128] After transmission is complete, the buffer is cleared to prepare for the next data collection.

[0129] Step 3:

[0130] Data Analysis

[0131] The server receives and analyzes data sent from the terminal. The input is location and activity data sent from the terminal, and the output is analyzed features (e.g., activity level, number of meals, frequency of toilet use, etc.) and anomaly detection results. First, the raw data is converted into an analyzable format (e.g., CSV or JSON), and outliers and missing values ​​are checked and cleaned. Then, specific features are extracted, and abnormal behavior is analyzed using a machine learning algorithm.

[0132] Specific actions:

[0133] Convert the raw data to a different format and clean the data as needed.

[0134] Extract features and organize them into an analyzable format.

[0135] Machine learning algorithms are applied to detect anomalies.

[0136] Step 4:

[0137] Abnormal notification

[0138] The server notifies the terminal of any anomalies detected. The input consists of the analyzed data and the results of the anomaly detection, while the output is the notification message sent to the terminal. The notification includes the type, date, time, and frequency of the detected anomaly.

[0139] Specific actions:

[0140] A notification message is constructed based on the detected anomaly data.

[0141] Send a message to the device and display it on the user interface.

[0142] Step 5:

[0143] Feedback Collection

[0144] The device collects user feedback and sends it to the server. The input is feedback information provided by the user, and the output is feedback data sent to the server. The feedback includes detailed information such as the user's observations of the pet's condition, the actions taken, and the results.

[0145] Specific actions:

[0146] Collect feedback through the user interface.

[0147] The collected feedback is sent to the server.

[0148] Step 6:

[0149] Model Retraining

[0150] The server retrains its machine learning algorithm based on feedback received from the user. The input consists of feedback data and historical behavioral data, and the output is an updated anomaly detection model. This improves the accuracy of anomaly detection.

[0151] Specific actions:

[0152] Integrate feedback data with past behavioral data.

[0153] Retrain the anomaly detection model and apply the new model.

[0154] (Application Example 1)

[0155] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0156] Properly monitoring a pet's health while it is traveling and detecting abnormalities early is crucial for ensuring their safety and well-being. Conventional pet health monitoring systems are typically used within the home or in specific locations, and monitoring during travel has not been considered. As a result, it has been difficult to ascertain a pet's health status in a timely manner, especially during long journeys. Furthermore, while a quick and accurate response is required when an abnormality is detected in a pet, there is a challenge in that appropriate means of response are not readily available while the pet is traveling.

[0157] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0158] In this invention, the server includes means for collecting pet behavior data using sensors and cameras, means for temporarily storing the collected behavior data, means for transmitting the stored behavior data to the server, means for analyzing the transmitted behavior data and applying a machine learning model to detect anomalies, means for notifying the terminal of the detected anomaly, means for displaying the notified anomaly information and providing the user with appropriate countermeasures, means for collecting user feedback and transmitting it to the server, means for retraining the machine learning model using the transmitted feedback, means for collecting pet behavior data using sensors and cameras mounted in the autonomous vehicle, means for processing the data in real time and detecting anomalies in cooperation with the system in the autonomous vehicle, and means for notifying the vehicle's dashboard display and smartphone when an anomaly is detected. This makes it possible to appropriately monitor the health status of pets even while they are in transit and to detect anomalies early.

[0159] "Pet behavioral data" refers to information related to a pet's movement patterns, activity level, and health status.

[0160] A "sensor" is a device that captures environmental information or physical quantities and converts them into electrical signals, etc. Specific examples include GPS sensors and accelerometers.

[0161] A "camera" is a device that captures images and videos, and includes 360-degree cameras and night vision cameras.

[0162] "Means of temporary storage" refers to memory or buffers used to retain data for a certain period of time.

[0163] "Means of sending to the server" refers to a communication module that has the function of transferring collected data to a computer in a remote location.

[0164] A "machine learning model" refers to an algorithm that uses past data to recognize patterns and then analyzes and predicts new data.

[0165] "Means for detecting anomalies" refers to a function that uses machine learning models to identify anomalies that deviate from normal behavioral patterns.

[0166] "Means of notifying the terminal" refers to communication devices or applications that have the function of transmitting anomaly detection information to the user.

[0167] A "dashboard display" refers to a display device installed inside an autonomous vehicle, used to visually display notification information and warnings.

[0168] A "smartphone" refers to a mobile device equipped with multiple functions, such as voice communication, data communication, and application execution.

[0169] "User feedback" refers to responses and information provided by users regarding the system's operation and notifications.

[0170] "Methods for processing data in real time" refer to functions that perform analysis and decision-making almost simultaneously with the moment data is collected.

[0171] An "autonomous vehicle" refers to a vehicle that operates autonomously without the intervention of a driver.

[0172] This invention relates to a system designed for the early detection of pet health issues and illnesses. This system uses multiple sensors and cameras to collect behavioral data for monitoring pet health within autonomous vehicles, and analyzes this data on a server to detect abnormalities. The specific configuration and operation of this system are described in detail below.

[0173] Data collection

[0174] The terminal (a platform installed inside the autonomous vehicle) collects behavioral data in real time via sensors and cameras attached to the pet. For example, a GPS sensor tracks the pet's movement patterns, and an accelerometer records the pet's activity level. Cameras installed inside the vehicle capture the pet's actions as video, and the pet's movements are captured as video data. This data is temporarily stored in a buffer within the terminal. Specifically, if the pet is frequently moving or active inside the vehicle, its movements are recorded with a 360-degree camera.

[0175] Data transmission

[0176] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (TLS) to ensure data consistency and security. The data is time-stamped to accurately preserve its order.

[0177] Data analysis and anomaly detection

[0178] The server analyzes the received data. First, it converts the data into a format that can be analyzed, and then cleans it by checking for outliers and missing values. Next, it extracts features to compare with past behavioral data. Specific features include exercise levels, meal frequency, and toilet usage frequency. Based on these features, a generative AI model is used to detect anomalies. When an abnormal behavioral pattern is detected, information that has been judged as abnormal is generated.

[0179] Anomaly notification and countermeasures provided

[0180] The server notifies the terminal of any detected anomalies. The terminal provides an interface to notify the user (pet owner) of this information, displaying it via voice, dashboard display, smartphone, etc. For example, if an anomaly is detected, an alert such as "Your pet is behaving unusually. Please check on them" will be displayed on the vehicle's dashboard display and smartphone.

[0181] User behavior and feedback

[0182] Based on the notified abnormal information, users can check their pet's condition and take appropriate action. For example, if the camera captures the pet frequently licking its front paws, the user can use that information to check the pet's paws and take it to the veterinarian if any abnormalities are found. Users can also provide feedback on their observations and the actions taken through the application.

[0183] Gathering feedback and retraining the model

[0184] The device sends feedback information provided by the user to the server. The server then retrains the anomaly detection model based on this feedback information, improving the system's accuracy. This makes it possible to monitor the pet's health more accurately.

[0185] As a concrete example, by using a prompt statement such as, "Write a program that analyzes sensor and camera data in real time to monitor the health of pets and detect abnormalities," the system's operation can be made to work exactly as specified.

[0186] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0187] Step 1:

[0188] The device collects behavioral data in real time using sensors and cameras attached to the pet. The input is raw data from the sensors and camera, and the output is data temporarily stored in a buffer within the device. Specifically, it simultaneously acquires the pet's location information from the GPS sensor, activity level from the accelerometer, and video data from the camera, and records each of these in the buffer.

[0189] Step 2:

[0190] The terminal sends data stored in a buffer to the server at regular intervals. The input here is the behavioral data stored in the buffer, and the output is the data sent to the server. The TLS protocol is used for this transmission, ensuring data consistency and security. Specifically, once data has accumulated in the buffer for a certain period, it is uploaded to the server in a single batch.

[0191] Step 3:

[0192] The server analyzes the received data. The input is behavioral data sent from the terminal, and the output is data converted into a parseable format. Specifically, after parsing the data, it checks for outliers and missing values ​​and performs cleaning. It also analyzes timestamp information to maintain the integrity of time-series data.

[0193] Step 4:

[0194] The server extracts features from the analyzed data and uses them to apply a generative AI model to detect anomalies. The input in this process is cleaned data, and the output is the anomaly detection result. Specifically, it extracts features such as exercise volume, meal frequency, and toilet usage frequency, and inputs these into the generative AI model to detect anomaly patterns.

[0195] Step 5:

[0196] The server notifies the terminal of any detected anomalies. The input is the result of the anomaly detection, and the output is the data for notification. Specifically, as soon as an anomaly is detected, the server sends that information to the terminal in real time and prepares to issue a warning to the user.

[0197] Step 6:

[0198] The terminal notifies the user of the received anomaly information. The input here is the anomaly information sent from the server, and the output is the notification to the user. Specifically, it displays the anomaly details and corrective actions using the vehicle's dashboard display or the smartphone's notification function.

[0199] Step 7:

[0200] The user checks the pet's condition based on the notified abnormal information and takes appropriate action as needed. The input is notifications from the terminal, and the output is user feedback. Specifically, the user actually observes the pet's condition and reports the results to the system via the terminal.

[0201] Step 8:

[0202] The terminal sends user-provided feedback information to the server. The input is the user's feedback, and the output is the data sent to the server. Specifically, it collects feedback information and sends it to the server along with a timestamp.

[0203] Step 9:

[0204] The server retrains the machine learning model using the feedback it receives. The input is the feedback data sent by the user, and the output is the new machine learning model with improved accuracy. Specifically, it incorporates the collected feedback data as part of the training data and updates the generative AI model.

[0205] By following the steps outlined above, it is possible to effectively implement a system that monitors a pet's health in real time and detects abnormalities early.

[0206] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0207] This invention is a system for supporting pet health management, which collects and analyzes pet behavioral data in real time and detects abnormalities. Furthermore, the invention incorporates an emotion engine that recognizes the user's emotions, and adds a function to adjust notifications and countermeasures based on the user's emotions. The specific configuration and embodiments of this system are described below.

[0208] Data collection

[0209] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. The GPS sensor tracks the pet's movement patterns, and the accelerometer records the pet's activity level. Simultaneously, a camera is used to capture the pet's behavior as video and detect specific actions (e.g., walking, sitting, licking). This data is temporarily stored in a buffer within the device.

[0210] Data transmission and analysis

[0211] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security.

[0212] The server converts the received data into an analyzable format, checks for outliers and missing values, and performs data cleaning. Next, it extracts features for comparison with past pet behavior data and uses a machine learning model to detect anomalies.

[0213] Using an Emotion Engine

[0214] This system includes an emotion engine that recognizes the user's emotions. The terminal captures the user's voice and facial expressions and sends that data to the server.

[0215] The server analyzes this data and applies emotion recognition algorithms. For example, it identifies whether the user is stressed or relaxed. This analysis result is taken into consideration when notifying users of abnormal information about their pet.

[0216] Anomaly notification and countermeasures provided

[0217] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the pet's abnormal behavior. The content and method of notification are adjusted based on the analysis results of the emotion engine. For example, if the user is stressed, the notification will be expressed in a softer tone, while if they are relaxed, more detailed information will be provided.

[0218] The device displays this information to the user through a user interface and provides appropriate countermeasures. For example, a message such as, "Your pet is behaving unusually. Please check on it," might be displayed.

[0219] User behavior and feedback

[0220] Users check notifications, observe their pet's condition, and take necessary actions. For example, they might check their pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken.

[0221] Gathering feedback and retraining the model

[0222] The device sends user-provided feedback information and sentiment data to the server.

[0223] Based on this feedback, the server retrains its anomaly detection model and emotion recognition algorithm to improve the system's accuracy and personalization. This allows for more accurate monitoring of pet health and the provision of optimal notifications and actions tailored to the user's emotional state.

[0224] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. Simultaneously, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and don't panic." The user can then use this information to check on their pet's condition and, if necessary, take it to a veterinarian, enabling early detection and treatment.

[0225] The following describes the processing flow.

[0226] Step 1:

[0227] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. Specifically, the GPS sensor periodically records the pet's location coordinates, and the accelerometer monitors the pet's activity level (e.g., intensity and frequency of movement). Furthermore, a camera is used to capture the pet's behavior as video, and specific actions (e.g., walking, sitting, licking) are automatically detected.

[0228] Step 2:

[0229] The device temporarily stores the collected behavioral data in a buffer. The buffer is time-stamped to accurately preserve the order of the data. For example, daily movement data, activity level data, and video data are stored in chronological order.

[0230] Step 3:

[0231] The terminal sends data stored in a buffer to the server at regular intervals (e.g., every minute). Data communication is securely performed using the TLS (Transport Layer Security) protocol, ensuring data consistency and security.

[0232] Step 4:

[0233] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and normalizes the sensor's numerical data. For example, the video data is converted into JPEG frames, and the acceleration data is scaled within a specified range.

[0234] Step 5:

[0235] The server performs data cleaning. It checks for outliers and missing values ​​and removes or fills in invalid data. For example, it identifies and processes abnormal spikes in activity level data or missing consecutive frames.

[0236] Step 6:

[0237] The server extracts features to compare with past pet behavior data stored in the database. For example, it calculates features such as daily exercise, number of meals, and frequency of litter box use.

[0238] Step 7:

[0239] The server uses a machine learning model to analyze features and detect abnormal pet behavior. For example, it can detect abnormal behavior such as "constantly licking its paws" by comparing it to normal behavior patterns. The machine learning model is trained based on past training data.

[0240] Step 8:

[0241] The device captures the user's voice and facial expressions and sends that data to a server. For example, it uses the smartphone's microphone and camera to record audio and video while the user is observing their pet.

[0242] Step 9:

[0243] The server analyzes the transmitted audio and video data and uses an emotion engine to recognize the user's emotions. For example, it can determine whether the user is stressed or relaxed based on changes in voice tone and facial expressions.

[0244] Step 10:

[0245] The server determines the content and method of notification based on the detected anomaly information and the results of the emotion engine's analysis. For example, if the user is feeling stressed, it will generate a gentle message such as, "Your pet is licking its paws repeatedly. We understand you are worried, but please take your time and check on it carefully."

[0246] Step 11:

[0247] The device displays the generated notification content to the user and suggests appropriate actions. For example, it might display a message via the smartphone's notification function saying, "Your pet is behaving unusually. We recommend consulting a veterinarian."

[0248] Step 12:

[0249] The user checks the notification, observes the pet's condition, and takes appropriate action (e.g., takes it to the vet). They then input feedback about the observation results and the actions taken into the application.

[0250] Step 13:

[0251] The device sends user-provided feedback and sentiment data to the server. For example, it sends text information entered by the user into the application and diagnostic results.

[0252] Step 14:

[0253] The server uses feedback information and sentiment data to retrain machine learning models and sentiment recognition algorithms, improving the system's accuracy. This allows the system to more accurately monitor pet health and provide optimal notifications and actions tailored to the user's emotional state.

[0254] (Example 2)

[0255] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0256] A current challenge in pet health management is the difficulty in quickly and accurately detecting abnormal behavior in pets and notifying users of this information in an appropriate manner. Furthermore, because the user's emotional state is not taken into consideration, the content of the notifications may cause excessive stress to the user. To solve these problems, this invention aims to provide a system that collects and analyzes not only pet behavior data but also user emotional data, and provides optimal notifications and countermeasures tailored to individual situations.

[0257] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0258] In this invention, the server includes means for analyzing pet behavior data and applying a machine learning model to detect anomalies, means for analyzing user emotion data and applying an emotion recognition algorithm, and means for adjusting anomaly notifications based on the user's emotional state. This enables high-precision detection of pet behavioral anomalies and the provision of appropriate notifications and countermeasures that take into account the user's emotional state.

[0259] "Pet behavioral data" refers to data that shows a pet's movement patterns, activity levels, and specific behaviors (e.g., walking, sitting, licking).

[0260] A "sensor" is a device used to measure the state of the environment or objects. Specifically, this includes GPS sensors and accelerometers.

[0261] A "camera" is a device used to capture video data.

[0262] A "buffer" is a memory area used to temporarily store data.

[0263] A "server" is a computer system that processes data over a network.

[0264] A "machine learning model" is a model built using algorithms to make predictions and classifications based on data.

[0265] "Emotional data" refers to data that indicates the emotional state of a user, obtained from their voice and facial expressions.

[0266] An "emotion identification algorithm" is an algorithm that analyzes emotional data to identify a user's emotional state (e.g., stress, relaxation).

[0267] An "abnormality notification" is a notification sent to the user when an abnormality is detected in the pet's behavior.

[0268] "User feedback" refers to information about observations and actions taken by users that they provide to the system.

[0269] "Retraining" is the process of improving a machine learning model using newly collected data.

[0270] This invention is a system for supporting pet health management, specifically by collecting and analyzing pet behavioral data in real time and detecting abnormalities. Furthermore, the invention incorporates an emotion engine that recognizes the user's emotions, and adds a function to adjust notifications and countermeasures based on the user's emotions.

[0271] Data collection

[0272] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. The GPS sensor tracks the pet's movement patterns, and the accelerometer records the pet's activity level. Simultaneously, a camera is used to capture the pet's behavior as video and detect specific actions (e.g., walking, sitting, licking). This data is temporarily stored in a buffer within the device.

[0273] Data transmission and analysis

[0274] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security.

[0275] The server converts the received data into an analyzable format, checks for outliers and missing values, and performs data cleaning. Next, it extracts features for comparison with past pet behavior data and uses a machine learning model to detect anomalies.

[0276] Using an Emotion Engine

[0277] This system includes an emotion engine that recognizes the user's emotions. The terminal captures the user's voice and facial expressions and sends that data to the server.

[0278] The server analyzes this data and applies emotion recognition algorithms. For example, it identifies whether the user is stressed or relaxed. This analysis is taken into consideration when notifying users of abnormal information about their pet.

[0279] Anomaly notification and countermeasures provided

[0280] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the pet's abnormal behavior. The content and method of notification are adjusted based on the analysis results of the emotion engine. For example, if the user is stressed, the notification will be expressed in a softer tone, while if they are relaxed, more detailed information will be provided.

[0281] The device displays this information to the user through a user interface and provides appropriate countermeasures. For example, a message such as, "Your pet is behaving unusually. Please check on it," might be displayed.

[0282] User behavior and feedback

[0283] The user checks the notification, actually observes the pet's condition, and takes necessary measures. For example, the user checks the pet's feet and consults a veterinarian if any abnormalities are found. Then, through the application, the user provides feedback on the observation results and the measures taken.

[0284] Collection of Feedback and Retraining of the Model

[0285] The terminal sends the feedback information and emotion data provided by the user to the server.

[0286] Based on this feedback information, the server retrains the anomaly detection model and the emotion recognition algorithm to improve the accuracy and personalization of the system. As a result, it becomes possible to more accurately monitor the pet's health condition and provide optimal notifications and countermeasures according to the user's emotional situation.

[0287] Specific Example

[0288] For example, consider the case where a camera captures a pet frequently licking its front paw. This video data is sent from the terminal to the server, and the server analyzes the behavior pattern of "continuously licking the paw" and determines it as an anomaly. At the same time, if the emotion engine detects that the user is feeling stressed from their voice and expression, the anomaly notification will include a gentle message such as "The pet is continuously licking its paw. There may be an abnormality in the paw. It is recommended to consult a veterinarian. However, let's stay calm and respond without getting anxious now." The user can refer to this and check the pet's condition, and if necessary, take the pet to the veterinarian, enabling early detection and early treatment.

[0289] Examples of Prompt Sentences:

[0290] "Please explain the mechanism of a system that detects abnormal behavior of a pet and provides appropriate notifications and countermeasures based on the user's emotions."

[0291] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0292] Step 1:

[0293] Data collection

[0294] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. First, it obtains the pet's current location from the GPS sensor every second, and measures the acceleration along the X, Y, and Z axes in milliseconds using the accelerometer. Furthermore, it captures the pet's behavior as video at 30 frames per second using a camera, and uses an image recognition algorithm to identify specific actions. This data is temporarily stored in a buffer. Sensor data and video data are collected as input and stored in the buffer as output.

[0295] Step 2:

[0296] Data transmission

[0297] The device sends activity data stored in a buffer to the server every 30 seconds. The transmitted data includes GPS data, acceleration data, and video data. The transmission process uses the TLS protocol to ensure data security and consistency. It takes data from the buffer as input and sends it to the server as output.

[0298] Step 3:

[0299] Data Analysis

[0300] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and identifies the pet's movements using an image recognition algorithm. GPS data and acceleration data are synchronized using timestamps, and outliers and missing values ​​are filtered out. Features are extracted using the cleaned data. The server takes the received data as input and obtains the analyzed data and features as output.

[0301] Step 4:

[0302] Collection and Transmission of Emotional Data

[0303] The terminal uses an in-camera and a microphone to collect the user's emotional data. The camera captures the user's expressions, and the microphone records the voice. This data is temporarily stored in a buffer and sent to the server at regular intervals. The user's facial expression data and voice data are obtained as input, stored in the buffer as output, and sent to the server.

[0304] Step 5:

[0305] Emotion Analysis

[0306] The server analyzes the received emotional data. The facial recognition algorithm analyzes the feature points of the user's face to identify emotions. Also, the voice recognition algorithm analyzes the voice data to estimate emotions from the tone and speed of the voice. Using these analysis results, it is determined whether the user is feeling stressed or relaxed. The received emotional data is obtained as input, and the emotional state is obtained as output.

[0307] Step 6:

[0308] Anomaly Detection

[0309] The server detects anomalies using a machine learning model based on feature quantities. For example, if the pet is continuously performing the same action within a specific range (e.g., licking its front paw continuously), this is detected as an anomaly. The types of anomalies are classified and associated with specific behavior patterns. The cleaned data and feature quantities are obtained as input, and the anomaly detection results are obtained as output.

[0310] Step 7:

[0311] Anomaly Notification

[0312] The server generates a notification message based on the anomaly detection results. This message includes the type of abnormal behavior, date and time, frequency, and content based on the sentiment analysis results. It provides gentle information if the user is stressed, and detailed information if they are relaxed. The terminal displays the received notification to the user. For example, it displays the notification content using smartphone push notifications or the interface of a dedicated app. It takes anomaly detection results and sentiment analysis results as input, and displays the generated notification to the user as output.

[0313] Step 8:

[0314] Collecting user feedback

[0315] The user checks the notification and actually observes the pet's condition. For example, they check the pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken. The application takes the notification content as input and provides information regarding observations and actions taken as output.

[0316] Step 9:

[0317] Feedback analysis and model retraining

[0318] The device sends user-provided feedback and sentiment data to the server. The server uses this feedback to retrain its anomaly detection model and sentiment recognition algorithm. Specifically, it uses the feedback information as labeled data to improve the accuracy and personalization of the machine learning model. It retrieves the feedback information sent as input and obtains the improved anomaly detection model and sentiment recognition algorithm as output.

[0319] (Application Example 2)

[0320] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0321] Conventional pet health management systems have the functionality to collect and analyze pet behavior data in real time, but they do not provide notifications or countermeasures that take into account the user's emotional state. As a result, when the user is stressed, the notification content may be inappropriate, and there may be cases where an abnormality in the pet cannot be addressed quickly. The present invention aims to provide a more effective pet health management system that takes the user's emotions into consideration.

[0322] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting and transmitting user emotion data to the server, means for analyzing the transmitted emotion data and applying an emotion identification algorithm, and means for adjusting countermeasures based on the detected anomaly information and the user's emotional state and notifying the terminal. This makes it possible to monitor the pet's health status in real time while providing appropriate notifications and countermeasures according to the user's emotional state.

[0323] "Behavioral data" refers to data that shows a pet's movement patterns and activity levels, and is collected using GPS sensors and accelerometers.

[0324] A "sensor" is a device that detects physical information about the environment or objects, such as a GPS sensor or an accelerometer, and outputs that information as data.

[0325] A "camera" is a device used to capture a pet's behavior as video, and it is equipped with the function to detect specific actions.

[0326] A "server" refers to a computer system used to analyze collected data, detect anomalies, and analyze user emotions.

[0327] A "machine learning model" includes data analysis algorithms used to detect anomalies based on pet behavior data.

[0328] "Emotion recognition" refers to an algorithm that identifies a user's emotional state by analyzing their voice and facial expression data.

[0329] An "abnormal" behavior refers to a state in which a pet exhibits behavior that is clearly different from its normal behavioral patterns, suggesting a potential health problem.

[0330] "Notifications" refer to information sent to inform users about detected anomalies or measures taken based on the user's emotional state.

[0331] "Countermeasures" include instructions that show how to respond to detected anomalies or what actions to take.

[0332] "User feedback" refers to information provided by users regarding their pet's condition and the measures taken, which is used to improve the system's performance.

[0333] "Retraining" refers to the process of improving the performance of machine learning models and sentiment recognition algorithms using newly collected data.

[0334] This invention is a system for supporting pet health management, which uses sensors and cameras to collect and analyze pet behavior data in real time. Furthermore, by collecting and analyzing user emotional data, the system can provide notifications and countermeasures based on the user's emotional state. Based on this embodiment, a method for effectively managing a pet's health will be described.

[0335] System Configuration

[0336] The system mainly consists of the following components:

[0337] Sensors: These are GPS sensors and accelerometers attached to the pet, which are used to obtain the pet's location information and movement data.

[0338] Camera: A high-resolution camera that captures specific actions of your pet (e.g., walking, sitting, licking).

[0339] Devices (smartphones, tablets, etc.): Temporarily store collected data and send it to the server as needed.

[0340] Server: A computer system that performs data analysis, analyzing collected data using machine learning models and sentiment recognition algorithms.

[0341] User interface: An application used to notify users of abnormal information and countermeasures.

[0342] System operation

[0343] 1. Data Collection: Pet behavior data collected by sensors (e.g., GPS data, accelerometer data) and behavioral video data captured by cameras are transmitted to the device. The device also collects user voice and facial expression data.

[0344] 2. Data transmission: The collected data is sent to the server at regular intervals using a secure protocol.

[0345] 3. Analysis and Anomaly Detection: The server converts the received data into an analyzable format and uses machine learning models to detect abnormal pet behavior. Furthermore, it analyzes the user's emotional state using an emotion recognition algorithm.

[0346] 4. Provision of Notifications and Countermeasures: If an anomaly is detected, the server sends the most appropriate notification and countermeasure to the terminal based on the anomaly information and the user's emotional state. The terminal then notifies the user of this through the user interface.

[0347] 5. User Response and Feedback: Users will review notifications and take appropriate action regarding their pet's abnormal behavior. They will provide feedback on their observations and the measures taken through the application.

[0348] 6. System Improvement: The server retrains machine learning models and sentiment recognition algorithms based on user feedback to improve the system's accuracy and personalization.

[0349] Specific example

[0350] For example, if the camera captures a pet frequently licking its front paws, the video data is sent from the device to the server. The server analyzes the data and determines that the "continuous paw licking" behavior is abnormal. At the same time, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and do not panic." This allows the user to take appropriate action and effectively manage their pet's health.

[0351] Example of a prompt

[0352] "What should I do if my pet is frequently exhibiting signs of illness?"

[0353] This system allows for real-time monitoring of pet health and prompt, appropriate action. Furthermore, by providing notifications and solutions tailored to the user's emotional state, it reduces the user's mental burden, enabling more confident pet health management.

[0354] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0355] Step 1:

[0356] The device receives behavioral data collected by GPS and accelerometer sensors attached to the pet. It also collects video data of the pet's behavior captured by a camera. Simultaneously, the device collects the user's voice and facial expression data. In this step, the inputs are data from sensors and cameras, and the user's voice and facial expression data, and the output is raw data that is temporarily stored.

[0357] Step 2:

[0358] The terminal sends the collected data to the server at regular intervals. Communication is performed using a secure protocol (e.g., TLS). In this step, there is temporarily stored data as input, and data is generated to be sent to the server as output.

[0359] Step 3:

[0360] The server converts the received data into a parseable format. It performs data cleaning, checking for outliers and missing values, and formatting the data. In this step, the raw data received is the input, and the cleaned data is produced as the output.

[0361] Step 4:

[0362] The server inputs the formatted data into a machine learning model to analyze the pet's behavior patterns. An anomaly detection model is applied to detect abnormal behavior. In this step, the input is cleaned data, and the output is the anomaly detection result.

[0363] Step 5:

[0364] The server analyzes the user's voice and facial expression data and applies an emotion recognition algorithm to determine whether the user is stressed or relaxed. In this step, the user's voice and facial expression data are the input, and the emotion recognition result is generated as the output.

[0365] Step 6:

[0366] The server integrates anomaly detection results and sentiment identification results to generate appropriate notification content and countermeasures. It adjusts the text and wording of the notification based on the user's emotional state. In this step, the input is anomaly detection results and sentiment identification results, and the output is the adjusted notification content.

[0367] Step 7:

[0368] The terminal receives notifications from the server and displays them through the user interface. The user reviews the notification content and takes appropriate action. In this step, the input is the adjusted notification content, and the output is the notification sent to the user.

[0369] Step 8:

[0370] The user checks the pet's condition and takes necessary actions based on the notification. Feedback on observations and actions taken is provided through the application. In this step, the user's observation and action information is the input, and feedback information is generated as the output.

[0371] Step 9:

[0372] The server receives feedback from the user and uses it to retrain the machine learning model and sentiment recognition algorithm. In this step, the received feedback information is taken as input, and the updated model is produced as output.

[0373] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0374] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0375] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0376] [Second Embodiment]

[0377] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0378] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0379] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0380] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0381] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0382] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0383] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0384] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0385] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0386] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0387] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0388] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0389] This invention is a system designed to detect the health status and illnesses of pets at an early stage. This system collects behavioral data from pets using multiple sensors and cameras, and analyzes this data on a server to detect abnormalities. The specific configuration and operation of this system are described in detail below.

[0390] Data collection

[0391] The device collects behavioral data in real time via sensors and cameras attached to the pet. For example, it tracks the pet's movement patterns with a GPS sensor and records its activity level with an accelerometer. It captures the pet's behavior as video using a camera, and records the pet's movements as video data. This data is temporarily stored in a buffer within the device.

[0392] Data transmission

[0393] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security. The data is time-stamped to maintain its order precisely.

[0394] Data analysis and anomaly detection

[0395] The server analyzes the received data. First, it converts the data into a format that can be analyzed, and then cleans it by checking for outliers and missing values. Next, it extracts features to compare with past behavioral data. Specific features include exercise levels, meal frequency, and toilet usage frequency. Based on these features, a machine learning model is used to detect anomalies. When an abnormal behavioral pattern is detected, information indicating that the behavior is abnormal is generated.

[0396] Anomaly notification and countermeasures provided

[0397] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the detected abnormal behavior. The terminal displays this information to the user (pet owner) through a user interface and provides appropriate countermeasures. For example, if an anomaly is detected, an alert such as "Your pet is behaving unusually. Please check on them" is displayed to the user.

[0398] User behavior and feedback

[0399] Based on the notified abnormal information, the user checks the pet's current condition and takes appropriate action as needed. The user observes the pet's condition and takes action, such as taking it to a veterinarian. Afterwards, they provide feedback through the application regarding the actual observation results and the actions taken.

[0400] Gathering feedback and retraining the model

[0401] The device sends feedback information provided by the user to the server. The server then uses this feedback information to retrain the anomaly detection model, improving the system's accuracy. This allows for more accurate monitoring of the pet's health.

[0402] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. The device then sends a notification to the user stating, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian." Based on this information, the user can check their pet's paws and, if there is a problem, take it to a veterinarian, enabling early detection and treatment.

[0403] The following describes the processing flow.

[0404] Step 1:

[0405] The device collects behavioral data in real time using sensors and cameras attached to the pet. Specifically, it tracks movement patterns with a GPS sensor and records activity levels with an accelerometer. At the same time, it captures behavior as video with a camera and detects specific actions (e.g., walking, sitting, licking).

[0406] Step 2:

[0407] The device temporarily stores the collected data in a buffer. The buffer is time-stamped to accurately preserve the order of the data. For example, movement data, activity level data, and video data are stored in chronological order.

[0408] Step 3:

[0409] The terminal sends data from a buffer to the server at regular intervals. This transmission process uses secure protocols such as TLS to ensure data consistency and security. For example, data batches are sent to the server every minute.

[0410] Step 4:

[0411] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and normalizes the sensor's numerical data. For example, it converts the video data into JPEG frames and scales the acceleration data within a specified range.

[0412] Step 5:

[0413] The server performs data cleaning. It checks for outliers (e.g., obviously incorrect numbers or corrupted data) and missing values, and removes or fills in the invalid data. For example, it might fill in abnormal spikes in activity level data or missing consecutive frames.

[0414] Step 6:

[0415] The server extracts features to compare with past pet behavior data stored in the database. For example, it calculates features such as daily exercise, number of meals, and frequency of litter box use.

[0416] Step 7:

[0417] The server uses a machine learning model to analyze features and detect anomalies. Specifically, it compares abnormal behavior (e.g., continuous licking of paws, sudden decrease in activity level) with normal behavior patterns to detect abnormal behavior. The model is trained based on past data and feedback.

[0418] Step 8:

[0419] The server notifies the terminal of the detected anomaly. The anomaly information includes details such as the type of abnormal behavior detected, the date and time, and the frequency.

[0420] Step 9:

[0421] The device notifies the user of any abnormal information it receives via the user interface. For example, a smartphone application might display a pop-up notification with a message such as, "Your pet is behaving unusually. Please check on it."

[0422] Step 10:

[0423] The user checks the notification, observes their pet's condition, and takes necessary action. For example, they might check their pet's paws and consult a veterinarian if any abnormalities are found.

[0424] Step 11:

[0425] Users provide feedback through the application, including observations and actions taken. Specifically, they input text information about their pet's current condition, as well as veterinary diagnoses.

[0426] Step 12:

[0427] The device sends user feedback information to the server. The transmitted feedback includes specific observations and details of treatment.

[0428] Step 13:

[0429] The server analyzes the received feedback information and retrains the machine learning model based on the new insights. This improves the system's anomaly detection accuracy and provides more precise diagnoses.

[0430] (Example 1)

[0431] Next, we will describe Example 1. 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".

[0432] In modern society, it is extremely important to monitor pets' health in real time and detect abnormalities early. However, with conventional systems, pet owners often fail to notice their pets' illnesses until it is too late, making early treatment difficult. Furthermore, there is a lack of efficient systems for collecting and analyzing pet behavioral data and developing appropriate countermeasures based on that data.

[0433] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0434] In this invention, the server includes means for collecting pet behavior data using a detection device, means for temporarily storing the collected behavior data, means for transmitting the stored behavior data to an information processing device, means for analyzing the transmitted behavior data and applying a machine learning algorithm to detect anomalies, means for notifying a display terminal of the detected anomaly information, means for displaying the notified anomaly information and providing the user with appropriate countermeasures, means for collecting user feedback and transmitting it to the information processing device, and means for retraining the machine learning algorithm using the transmitted feedback. This enables real-time monitoring of the pet's health status, early detection of anomalies, and prompt and appropriate responses.

[0435] A "detection device" is a device that includes sensors and cameras used to collect pet behavioral data.

[0436] A "location detection sensor" is a sensor used to obtain a pet's location information, and specifically includes GPS sensors.

[0437] A "activity detection sensor" is a sensor used to measure a pet's activity level, and specifically includes accelerometers and similar sensors.

[0438] "Behavioral data" refers to data including the pet's location, movements, and activity level.

[0439] "Means of temporary storage" refers to memory or storage for temporarily storing collected data.

[0440] An "information processing device" is a computing device that analyzes received data and outputs the results.

[0441] A "machine learning algorithm" is a mathematical model used to detect anomalies based on past data.

[0442] A "display terminal" is a device used to notify users of abnormal information transmitted from a server. Specifically, this includes smartphones and tablets.

[0443] "Users" refer to pet owners or caretakers.

[0444] "Feedback" refers to information about what users observed and the actions they took.

[0445] This invention relates to a system for early detection of the health status or illness of pets. This system uses multiple hardware and software components to collect and analyze pet behavioral data and detect abnormalities. Specifically, it uses a combination of sensors and cameras as detection devices, an information processing device for analysis and notification, and a terminal for collecting feedback.

[0446] The device collects real-time behavioral data of the pet using location detection sensors (e.g., GPS sensors) and activity detection sensors (e.g., accelerometers) attached to the pet. This data is stored in the device's temporary storage. The collected data is analyzed and notifications are sent according to the following processing flow.

[0447] Hardware and software usage

[0448] 1. Detection device:

[0449] Location sensors collect information about a pet's current location. For example, a GPS sensor tracks the pet's movement patterns.

[0450] The activity detection sensor records the pet's activity level and uses an accelerometer to measure how active the pet is.

[0451] 2. Information processing device:

[0452] Data transmission: The terminal sends the collected data to the server (information processing device) at regular intervals (e.g., every 5 minutes). This process is carried out using a secure protocol (e.g., TLS).

[0453] Data Analysis and Anomaly Detection: The server analyzes the received data and detects anomalies. First, it converts the data into a parseable format (e.g., CSV or JSON). Next, it cleans the data and extracts features (such as exercise levels, meal frequency, and toilet usage frequency). Based on these features, it applies machine learning algorithms to detect abnormal behavior.

[0454] 3. Notice:

[0455] Anomaly Notification: The server notifies the terminal of any detected anomalies. The terminal displays this information to the user through a user interface (such as a smartphone app). For example, an alert might be displayed stating, "Your pet is behaving unusually. Please check on it."

[0456] 4. Gathering feedback:

[0457] User Behavior and Feedback: Users check the status of their pets based on the notified abnormal information and take appropriate measures as needed. Users provide feedback on their observations and actions taken through the application.

[0458] Model retraining: The terminal sends the provided feedback information to the server, which then retrains the anomaly detection model based on this feedback. This improves the system's accuracy and enables more precise anomaly detection in the future.

[0459] Specific example

[0460] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws, and this video data is sent from the device to a server. The server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. As a result, the device sends a notification to the user stating, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian." Based on this information, the user can check their pet's paws and, if there is a problem, take it to a veterinarian, enabling early detection and treatment.

[0461] An example of a prompt statement is: "Use a system that collects pet behavior data using cameras and sensors and analyzes it on a server to monitor the pet's health in real time and detect abnormalities."

[0462] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0463] Step 1:

[0464] Data collection

[0465] The device collects behavioral data in real time using location sensors (e.g., GPS sensors) and activity sensors (e.g., accelerometers) attached to the pet. Inputs include signals from the sensors (location information, acceleration data). These sensor signals are converted into digital data and stored in a temporary buffer memory within the device. Outputs consist of a set of location information and activity data.

[0466] Specific actions:

[0467] The GPS sensor collects location information every second.

[0468] The accelerometer records the pet's movements every 0.1 seconds.

[0469] The collected data is temporarily stored in a buffer.

[0470] Step 2:

[0471] Data transmission

[0472] The terminal sends data stored in a buffer to the server at regular intervals (e.g., every 5 minutes). The input is location and momentum data stored in the buffer, and the output is data packets sent to the server. Secure protocols such as TLS are used for transmission, and the order of the data is maintained by timestamping each data packet.

[0473] Specific actions:

[0474] The data stored in the buffer is read and sent to the server using the TLS protocol.

[0475] After transmission is complete, the buffer is cleared to prepare for the next data collection.

[0476] Step 3:

[0477] Data Analysis

[0478] The server receives and analyzes data sent from the terminal. The input is location and activity data sent from the terminal, and the output is analyzed features (e.g., activity level, number of meals, frequency of toilet use, etc.) and anomaly detection results. First, the raw data is converted into an analyzable format (e.g., CSV or JSON), and outliers and missing values ​​are checked and cleaned. Then, specific features are extracted, and abnormal behavior is analyzed using a machine learning algorithm.

[0479] Specific actions:

[0480] Convert the raw data to a different format and clean the data as needed.

[0481] Extract features and organize them into an analyzable format.

[0482] Machine learning algorithms are applied to detect anomalies.

[0483] Step 4:

[0484] Abnormal notification

[0485] The server notifies the terminal of any anomalies detected. The input consists of the analyzed data and the results of the anomaly detection, while the output is the notification message sent to the terminal. The notification includes the type, date, time, and frequency of the detected anomaly.

[0486] Specific actions:

[0487] A notification message is constructed based on the detected anomaly data.

[0488] Send a message to the device and display it on the user interface.

[0489] Step 5:

[0490] Feedback Collection

[0491] The device collects user feedback and sends it to the server. The input is feedback information provided by the user, and the output is feedback data sent to the server. The feedback includes detailed information such as the user's observations of the pet's condition, the actions taken, and the results.

[0492] Specific actions:

[0493] Collect feedback through the user interface.

[0494] The collected feedback is sent to the server.

[0495] Step 6:

[0496] Model Retraining

[0497] The server retrains its machine learning algorithm based on feedback received from the user. The input consists of feedback data and historical behavioral data, and the output is an updated anomaly detection model. This improves the accuracy of anomaly detection.

[0498] Specific actions:

[0499] Integrate feedback data with past behavioral data.

[0500] Retrain the anomaly detection model and apply the new model.

[0501] (Application Example 1)

[0502] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0503] Properly monitoring a pet's health while it is traveling and detecting abnormalities early is crucial for ensuring their safety and well-being. Conventional pet health monitoring systems are typically used within the home or in specific locations, and monitoring during travel has not been considered. As a result, it has been difficult to ascertain a pet's health status in a timely manner, especially during long journeys. Furthermore, while a quick and accurate response is required when an abnormality is detected in a pet, there is a challenge in that appropriate means of response are not readily available while the pet is traveling.

[0504] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0505] In this invention, the server includes means for collecting pet behavior data using sensors and cameras, means for temporarily storing the collected behavior data, means for transmitting the stored behavior data to the server, means for analyzing the transmitted behavior data and applying a machine learning model to detect anomalies, means for notifying the terminal of the detected anomaly, means for displaying the notified anomaly information and providing the user with appropriate countermeasures, means for collecting user feedback and transmitting it to the server, means for retraining the machine learning model using the transmitted feedback, means for collecting pet behavior data using sensors and cameras mounted in the autonomous vehicle, means for processing the data in real time and detecting anomalies in cooperation with the system in the autonomous vehicle, and means for notifying the vehicle's dashboard display and smartphone when an anomaly is detected. This makes it possible to appropriately monitor the health status of pets even while they are in transit and to detect anomalies early.

[0506] "Pet behavioral data" refers to information related to a pet's movement patterns, activity level, and health status.

[0507] A "sensor" is a device that captures environmental information or physical quantities and converts them into electrical signals, etc. Specific examples include GPS sensors and accelerometers.

[0508] A "camera" is a device that captures images and videos, and includes 360-degree cameras and night vision cameras.

[0509] "Means of temporary storage" refers to memory or buffers used to retain data for a certain period of time.

[0510] "Means of sending to the server" refers to a communication module that has the function of transferring collected data to a computer in a remote location.

[0511] A "machine learning model" refers to an algorithm that uses past data to recognize patterns and then analyzes and predicts new data.

[0512] "Means for detecting anomalies" refers to a function that uses machine learning models to identify anomalies that deviate from normal behavioral patterns.

[0513] "Means of notifying the terminal" refers to communication devices or applications that have the function of transmitting anomaly detection information to the user.

[0514] A "dashboard display" refers to a display device installed inside an autonomous vehicle, used to visually display notification information and warnings.

[0515] A "smartphone" refers to a mobile device equipped with multiple functions, such as voice communication, data communication, and application execution.

[0516] "User feedback" refers to responses and information provided by users regarding the system's operation and notifications.

[0517] "Methods for processing data in real time" refer to functions that perform analysis and decision-making almost simultaneously with the moment data is collected.

[0518] An "autonomous vehicle" refers to a vehicle that operates autonomously without the intervention of a driver.

[0519] This invention relates to a system designed for the early detection of pet health issues and illnesses. This system uses multiple sensors and cameras to collect behavioral data for monitoring pet health within autonomous vehicles, and analyzes this data on a server to detect abnormalities. The specific configuration and operation of this system are described in detail below.

[0520] Data collection

[0521] The terminal (a platform installed inside the autonomous vehicle) collects behavioral data in real time via sensors and cameras attached to the pet. For example, a GPS sensor tracks the pet's movement patterns, and an accelerometer records the pet's activity level. Cameras installed inside the vehicle capture the pet's actions as video, and the pet's movements are captured as video data. This data is temporarily stored in a buffer within the terminal. Specifically, if the pet is frequently moving or active inside the vehicle, its movements are recorded with a 360-degree camera.

[0522] Data transmission

[0523] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (TLS) to ensure data consistency and security. The data is time-stamped to accurately preserve its order.

[0524] Data analysis and anomaly detection

[0525] The server analyzes the received data. First, it converts the data into a format that can be analyzed, and then cleans it by checking for outliers and missing values. Next, it extracts features to compare with past behavioral data. Specific features include exercise levels, meal frequency, and toilet usage frequency. Based on these features, a generative AI model is used to detect anomalies. When an abnormal behavioral pattern is detected, information that has been judged as abnormal is generated.

[0526] Anomaly notification and countermeasures provided

[0527] The server notifies the terminal of any detected anomalies. The terminal provides an interface to notify the user (pet owner) of this information, displaying it via voice, dashboard display, smartphone, etc. For example, if an anomaly is detected, an alert such as "Your pet is behaving unusually. Please check on them" will be displayed on the vehicle's dashboard display and smartphone.

[0528] User behavior and feedback

[0529] Based on the notified abnormal information, users can check their pet's condition and take appropriate action. For example, if the camera captures the pet frequently licking its front paws, the user can use that information to check the pet's paws and take it to the veterinarian if any abnormalities are found. Users can also provide feedback on their observations and the actions taken through the application.

[0530] Gathering feedback and retraining the model

[0531] The device sends feedback information provided by the user to the server. The server then retrains the anomaly detection model based on this feedback information, improving the system's accuracy. This makes it possible to monitor the pet's health more accurately.

[0532] As a concrete example, by using a prompt statement such as, "Write a program that analyzes sensor and camera data in real time to monitor the health of pets and detect abnormalities," the system's operation can be made to work exactly as specified.

[0533] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0534] Step 1:

[0535] The device collects behavioral data in real time using sensors and cameras attached to the pet. The input is raw data from the sensors and camera, and the output is data temporarily stored in a buffer within the device. Specifically, it simultaneously acquires the pet's location information from the GPS sensor, activity level from the accelerometer, and video data from the camera, and records each of these in the buffer.

[0536] Step 2:

[0537] The terminal sends data stored in a buffer to the server at regular intervals. The input here is the behavioral data stored in the buffer, and the output is the data sent to the server. The TLS protocol is used for this transmission, ensuring data consistency and security. Specifically, once data has accumulated in the buffer for a certain period, it is uploaded to the server in a single batch.

[0538] Step 3:

[0539] The server analyzes the received data. The input is behavioral data sent from the terminal, and the output is data converted into a parseable format. Specifically, after parsing the data, it checks for outliers and missing values ​​and performs cleaning. It also analyzes timestamp information to maintain the integrity of time-series data.

[0540] Step 4:

[0541] The server extracts features from the analyzed data and uses them to apply a generative AI model to detect anomalies. The input in this process is cleaned data, and the output is the anomaly detection result. Specifically, it extracts features such as exercise volume, meal frequency, and toilet usage frequency, and inputs these into the generative AI model to detect anomaly patterns.

[0542] Step 5:

[0543] The server notifies the terminal of any detected anomalies. The input is the result of the anomaly detection, and the output is the data for notification. Specifically, as soon as an anomaly is detected, the server sends that information to the terminal in real time and prepares to issue a warning to the user.

[0544] Step 6:

[0545] The terminal notifies the user of the received anomaly information. The input here is the anomaly information sent from the server, and the output is the notification to the user. Specifically, it displays the anomaly details and corrective actions using the vehicle's dashboard display or the smartphone's notification function.

[0546] Step 7:

[0547] The user checks the pet's condition based on the notified abnormal information and takes appropriate action as needed. The input is notifications from the terminal, and the output is user feedback. Specifically, the user actually observes the pet's condition and reports the results to the system via the terminal.

[0548] Step 8:

[0549] The terminal sends user-provided feedback information to the server. The input is the user's feedback, and the output is the data sent to the server. Specifically, it collects feedback information and sends it to the server along with a timestamp.

[0550] Step 9:

[0551] The server retrains the machine learning model using the feedback it receives. The input is the feedback data sent by the user, and the output is the new machine learning model with improved accuracy. Specifically, it incorporates the collected feedback data as part of the training data and updates the generative AI model.

[0552] By following the steps outlined above, it is possible to effectively implement a system that monitors a pet's health in real time and detects abnormalities early.

[0553] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0554] This invention is a system for supporting pet health management, which collects and analyzes pet behavioral data in real time and detects abnormalities. Furthermore, the invention incorporates an emotion engine that recognizes the user's emotions, and adds a function to adjust notifications and countermeasures based on the user's emotions. The specific configuration and embodiments of this system are described below.

[0555] Data collection

[0556] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. The GPS sensor tracks the pet's movement patterns, and the accelerometer records the pet's activity level. Simultaneously, a camera is used to capture the pet's behavior as video and detect specific actions (e.g., walking, sitting, licking). This data is temporarily stored in a buffer within the device.

[0557] Data transmission and analysis

[0558] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security.

[0559] The server converts the received data into an analyzable format, checks for outliers and missing values, and performs data cleaning. Next, it extracts features for comparison with past pet behavior data and uses a machine learning model to detect anomalies.

[0560] Using an Emotion Engine

[0561] This system includes an emotion engine that recognizes the user's emotions. The terminal captures the user's voice and facial expressions and sends that data to the server.

[0562] The server analyzes this data and applies emotion recognition algorithms. For example, it identifies whether the user is stressed or relaxed. This analysis result is taken into consideration when notifying users of abnormal information about their pet.

[0563] Anomaly notification and countermeasures provided

[0564] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the pet's abnormal behavior. The content and method of notification are adjusted based on the analysis results of the emotion engine. For example, if the user is stressed, the notification will be expressed in a softer tone, while if they are relaxed, more detailed information will be provided.

[0565] The device displays this information to the user through a user interface and provides appropriate countermeasures. For example, a message such as, "Your pet is behaving unusually. Please check on it," might be displayed.

[0566] User behavior and feedback

[0567] Users check notifications, observe their pet's condition, and take necessary actions. For example, they might check their pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken.

[0568] Gathering feedback and retraining the model

[0569] The device sends user-provided feedback information and sentiment data to the server.

[0570] Based on this feedback, the server retrains its anomaly detection model and emotion recognition algorithm to improve the system's accuracy and personalization. This allows for more accurate monitoring of pet health and the provision of optimal notifications and actions tailored to the user's emotional state.

[0571] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. Simultaneously, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and don't panic." The user can then use this information to check on their pet's condition and, if necessary, take it to a veterinarian, enabling early detection and treatment.

[0572] The following describes the processing flow.

[0573] Step 1:

[0574] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. Specifically, the GPS sensor periodically records the pet's location coordinates, and the accelerometer monitors the pet's activity level (e.g., intensity and frequency of movement). Furthermore, a camera is used to capture the pet's behavior as video, and specific actions (e.g., walking, sitting, licking) are automatically detected.

[0575] Step 2:

[0576] The device temporarily stores the collected behavioral data in a buffer. The buffer is time-stamped to accurately preserve the order of the data. For example, daily movement data, activity level data, and video data are stored in chronological order.

[0577] Step 3:

[0578] The terminal sends data stored in a buffer to the server at regular intervals (e.g., every minute). Data communication is securely performed using the TLS (Transport Layer Security) protocol, ensuring data consistency and security.

[0579] Step 4:

[0580] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and normalizes the sensor's numerical data. For example, the video data is converted into JPEG frames, and the acceleration data is scaled within a specified range.

[0581] Step 5:

[0582] The server performs data cleaning. It checks for outliers and missing values ​​and removes or fills in invalid data. For example, it identifies and processes abnormal spikes in activity level data or missing consecutive frames.

[0583] Step 6:

[0584] The server extracts features to compare with past pet behavior data stored in the database. For example, it calculates features such as daily exercise, number of meals, and frequency of litter box use.

[0585] Step 7:

[0586] The server uses a machine learning model to analyze features and detect abnormal pet behavior. For example, it can detect abnormal behavior such as "constantly licking its paws" by comparing it to normal behavior patterns. The machine learning model is trained based on past training data.

[0587] Step 8:

[0588] The device captures the user's voice and facial expressions and sends that data to a server. For example, it uses the smartphone's microphone and camera to record audio and video while the user is observing their pet.

[0589] Step 9:

[0590] The server analyzes the transmitted audio and video data and uses an emotion engine to recognize the user's emotions. For example, it can determine whether the user is stressed or relaxed based on changes in voice tone and facial expressions.

[0591] Step 10:

[0592] The server determines the content and method of notification based on the detected anomaly information and the results of the emotion engine's analysis. For example, if the user is feeling stressed, it will generate a gentle message such as, "Your pet is licking its paws repeatedly. We understand you are worried, but please take your time and check on it carefully."

[0593] Step 11:

[0594] The device displays the generated notification content to the user and suggests appropriate actions. For example, it might display a message via the smartphone's notification function saying, "Your pet is behaving unusually. We recommend consulting a veterinarian."

[0595] Step 12:

[0596] The user checks the notification, observes the pet's condition, and takes appropriate action (e.g., takes it to the vet). They then input feedback about the observation results and the actions taken into the application.

[0597] Step 13:

[0598] The device sends user-provided feedback and sentiment data to the server. For example, it sends text information entered by the user into the application and diagnostic results.

[0599] Step 14:

[0600] The server uses feedback information and sentiment data to retrain machine learning models and sentiment recognition algorithms, improving the system's accuracy. This allows the system to more accurately monitor pet health and provide optimal notifications and actions tailored to the user's emotional state.

[0601] (Example 2)

[0602] Next, we will describe Example 2. 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".

[0603] A current challenge in pet health management is the difficulty in quickly and accurately detecting abnormal behavior in pets and notifying users of this information in an appropriate manner. Furthermore, because the user's emotional state is not taken into consideration, the content of the notifications may cause excessive stress to the user. To solve these problems, this invention aims to provide a system that collects and analyzes not only pet behavior data but also user emotional data, and provides optimal notifications and countermeasures tailored to individual situations.

[0604] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0605] In this invention, the server includes means for analyzing pet behavior data and applying a machine learning model to detect anomalies, means for analyzing user emotion data and applying an emotion recognition algorithm, and means for adjusting anomaly notifications based on the user's emotional state. This enables high-precision detection of pet behavioral anomalies and the provision of appropriate notifications and countermeasures that take into account the user's emotional state.

[0606] "Pet behavioral data" refers to data that shows a pet's movement patterns, activity levels, and specific behaviors (e.g., walking, sitting, licking).

[0607] A "sensor" is a device used to measure the state of the environment or objects. Specifically, this includes GPS sensors and accelerometers.

[0608] A "camera" is a device used to capture video data.

[0609] A "buffer" is a memory area used to temporarily store data.

[0610] A "server" is a computer system that processes data over a network.

[0611] A "machine learning model" is a model built using algorithms to make predictions and classifications based on data.

[0612] "Emotional data" refers to data that indicates the emotional state of a user, obtained from their voice and facial expressions.

[0613] An "emotion identification algorithm" is an algorithm that analyzes emotional data to identify a user's emotional state (e.g., stress, relaxation).

[0614] An "abnormality notification" is a notification sent to the user when an abnormality is detected in the pet's behavior.

[0615] "User feedback" refers to information about observations and actions taken by users that they provide to the system.

[0616] "Retraining" is the process of improving a machine learning model using newly collected data.

[0617] This invention is a system for supporting pet health management, specifically by collecting and analyzing pet behavioral data in real time and detecting abnormalities. Furthermore, the invention incorporates an emotion engine that recognizes the user's emotions, and adds a function to adjust notifications and countermeasures based on the user's emotions.

[0618] Data collection

[0619] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. The GPS sensor tracks the pet's movement patterns, and the accelerometer records the pet's activity level. Simultaneously, a camera is used to capture the pet's behavior as video and detect specific actions (e.g., walking, sitting, licking). This data is temporarily stored in a buffer within the device.

[0620] Data transmission and analysis

[0621] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security.

[0622] The server converts the received data into an analyzable format, checks for outliers and missing values, and performs data cleaning. Next, it extracts features for comparison with past pet behavior data and uses a machine learning model to detect anomalies.

[0623] Using an Emotion Engine

[0624] This system includes an emotion engine that recognizes the user's emotions. The terminal captures the user's voice and facial expressions and sends that data to the server.

[0625] The server analyzes this data and applies emotion recognition algorithms. For example, it identifies whether the user is stressed or relaxed. This analysis is taken into consideration when notifying users of abnormal information about their pet.

[0626] Anomaly notification and countermeasures provided

[0627] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the pet's abnormal behavior. The content and method of notification are adjusted based on the analysis results of the emotion engine. For example, if the user is stressed, the notification will be expressed in a softer tone, while if they are relaxed, more detailed information will be provided.

[0628] The device displays this information to the user through a user interface and provides appropriate countermeasures. For example, a message such as, "Your pet is behaving unusually. Please check on it," might be displayed.

[0629] User behavior and feedback

[0630] Users check notifications, observe their pet's condition, and take necessary actions. For example, they might check their pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken.

[0631] Gathering feedback and retraining the model

[0632] The device sends user-provided feedback information and sentiment data to the server.

[0633] Based on this feedback, the server retrains its anomaly detection model and emotion recognition algorithm to improve the system's accuracy and personalization. This allows for more accurate monitoring of pet health and the provision of optimal notifications and actions tailored to the user's emotional state.

[0634] Specific example

[0635] For example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. Simultaneously, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and don't panic." The user can then use this information to check on their pet's condition and, if necessary, take it to a veterinarian, enabling early detection and treatment.

[0636] Example of a prompt:

[0637] "Please explain how the system works to detect abnormal pet behavior and provide appropriate notifications and countermeasures based on the user's emotions."

[0638] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0639] Step 1:

[0640] Data collection

[0641] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. First, it obtains the pet's current location from the GPS sensor every second, and measures the acceleration along the X, Y, and Z axes in milliseconds using the accelerometer. Furthermore, it captures the pet's behavior as video at 30 frames per second using a camera, and uses an image recognition algorithm to identify specific actions. This data is temporarily stored in a buffer. Sensor data and video data are collected as input and stored in the buffer as output.

[0642] Step 2:

[0643] Data transmission

[0644] The device sends activity data stored in a buffer to the server every 30 seconds. The transmitted data includes GPS data, acceleration data, and video data. The transmission process uses the TLS protocol to ensure data security and consistency. It takes data from the buffer as input and sends it to the server as output.

[0645] Step 3:

[0646] Data Analysis

[0647] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and identifies the pet's movements using an image recognition algorithm. GPS data and acceleration data are synchronized using timestamps, and outliers and missing values ​​are filtered out. Features are extracted using the cleaned data. The server takes the received data as input and obtains the analyzed data and features as output.

[0648] Step 4:

[0649] Collection and transmission of emotional data

[0650] The device uses the front camera and microphone to collect user emotion data. The camera captures the user's facial expressions, and the microphone records their voice. This data is temporarily stored in a buffer and sent to the server at regular intervals. User facial expression data and voice data are taken as input, stored in a buffer as output, and sent to the server.

[0651] Step 5:

[0652] Emotion analysis

[0653] The server analyzes the received emotional data. It uses a facial recognition algorithm to analyze the user's facial features and identify emotions. It also uses a speech recognition algorithm to analyze audio data and estimate emotions from voice tone and speed. Using these analysis results, it determines whether the user is stressed or relaxed. It takes the received emotional data as input and obtains the emotional state as output.

[0654] Step 6:

[0655] Anomaly detection

[0656] The server uses a machine learning model to detect anomalies based on features. For example, if a pet repeatedly performs the same action within a specific range (e.g., continuously licking its front paws), this is detected as an anomaly. The server classifies the type of anomaly and associates it with specific behavioral patterns. It takes cleaned data and features as input and obtains anomaly detection results as output.

[0657] Step 7:

[0658] Abnormal notification

[0659] The server generates a notification message based on the anomaly detection results. This message includes the type of abnormal behavior, date and time, frequency, and content based on the sentiment analysis results. It provides gentle information if the user is stressed, and detailed information if they are relaxed. The terminal displays the received notification to the user. For example, it displays the notification content using smartphone push notifications or the interface of a dedicated app. It takes anomaly detection results and sentiment analysis results as input, and displays the generated notification to the user as output.

[0660] Step 8:

[0661] Collecting user feedback

[0662] The user checks the notification and actually observes the pet's condition. For example, they check the pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken. The application takes the notification content as input and provides information regarding observations and actions taken as output.

[0663] Step 9:

[0664] Feedback analysis and model retraining

[0665] The device sends user-provided feedback and sentiment data to the server. The server uses this feedback to retrain its anomaly detection model and sentiment recognition algorithm. Specifically, it uses the feedback information as labeled data to improve the accuracy and personalization of the machine learning model. It retrieves the feedback information sent as input and obtains the improved anomaly detection model and sentiment recognition algorithm as output.

[0666] (Application Example 2)

[0667] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0668] Conventional pet health management systems have the functionality to collect and analyze pet behavior data in real time, but they do not provide notifications or countermeasures that take into account the user's emotional state. As a result, when the user is stressed, the notification content may be inappropriate, and there may be cases where an abnormality in the pet cannot be addressed quickly. The present invention aims to provide a more effective pet health management system that takes the user's emotions into consideration.

[0669] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting and transmitting user emotion data to the server, means for analyzing the transmitted emotion data and applying an emotion identification algorithm, and means for adjusting countermeasures based on the detected anomaly information and the user's emotional state and notifying the terminal. This makes it possible to monitor the pet's health status in real time while providing appropriate notifications and countermeasures according to the user's emotional state.

[0670] "Behavioral data" refers to data that shows a pet's movement patterns and activity levels, and is collected using GPS sensors and accelerometers.

[0671] A "sensor" is a device that detects physical information about the environment or objects, such as a GPS sensor or an accelerometer, and outputs that information as data.

[0672] A "camera" is a device used to capture a pet's behavior as video, and it is equipped with the function to detect specific actions.

[0673] A "server" refers to a computer system used to analyze collected data, detect anomalies, and analyze user emotions.

[0674] A "machine learning model" includes data analysis algorithms used to detect anomalies based on pet behavior data.

[0675] "Emotion recognition" refers to an algorithm that identifies a user's emotional state by analyzing their voice and facial expression data.

[0676] An "abnormal" behavior refers to a state in which a pet exhibits behavior that is clearly different from its normal behavioral patterns, suggesting a potential health problem.

[0677] "Notifications" refer to information sent to inform users about detected anomalies or measures taken based on the user's emotional state.

[0678] "Countermeasures" include instructions that show how to respond to detected anomalies or what actions to take.

[0679] "User feedback" refers to information provided by users regarding their pet's condition and the measures taken, which is used to improve the system's performance.

[0680] "Retraining" refers to the process of improving the performance of machine learning models and sentiment recognition algorithms using newly collected data.

[0681] This invention is a system for supporting pet health management, which uses sensors and cameras to collect and analyze pet behavior data in real time. Furthermore, by collecting and analyzing user emotional data, the system can provide notifications and countermeasures based on the user's emotional state. Based on this embodiment, a method for effectively managing a pet's health will be described.

[0682] System Configuration

[0683] The system mainly consists of the following components:

[0684] Sensors: These are GPS sensors and accelerometers attached to the pet, which are used to obtain the pet's location information and movement data.

[0685] Camera: A high-resolution camera that captures specific actions of your pet (e.g., walking, sitting, licking).

[0686] Devices (smartphones, tablets, etc.): Temporarily store collected data and send it to the server as needed.

[0687] Server: A computer system that performs data analysis, analyzing collected data using machine learning models and sentiment recognition algorithms.

[0688] User interface: An application used to notify users of abnormal information and countermeasures.

[0689] System operation

[0690] 1. Data Collection: Pet behavior data collected by sensors (e.g., GPS data, accelerometer data) and behavioral video data captured by cameras are transmitted to the device. The device also collects user voice and facial expression data.

[0691] 2. Data transmission: The collected data is sent to the server at regular intervals using a secure protocol.

[0692] 3. Analysis and Anomaly Detection: The server converts the received data into an analyzable format and uses machine learning models to detect abnormal pet behavior. Furthermore, it analyzes the user's emotional state using an emotion recognition algorithm.

[0693] 4. Provision of Notifications and Countermeasures: If an anomaly is detected, the server sends the most appropriate notification and countermeasure to the terminal based on the anomaly information and the user's emotional state. The terminal then notifies the user of this through the user interface.

[0694] 5. User Response and Feedback: Users will review notifications and take appropriate action regarding their pet's abnormal behavior. They will provide feedback on their observations and the measures taken through the application.

[0695] 6. System Improvement: The server retrains machine learning models and sentiment recognition algorithms based on user feedback to improve the system's accuracy and personalization.

[0696] Specific example

[0697] For example, if the camera captures a pet frequently licking its front paws, the video data is sent from the device to the server. The server analyzes the data and determines that the "continuous paw licking" behavior is abnormal. At the same time, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and do not panic." This allows the user to take appropriate action and effectively manage their pet's health.

[0698] Example of a prompt

[0699] "What should I do if my pet is frequently exhibiting signs of illness?"

[0700] This system allows for real-time monitoring of pet health and prompt, appropriate action. Furthermore, by providing notifications and solutions tailored to the user's emotional state, it reduces the user's mental burden, enabling more confident pet health management.

[0701] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0702] Step 1:

[0703] The device receives behavioral data collected by GPS and accelerometer sensors attached to the pet. It also collects video data of the pet's behavior captured by a camera. Simultaneously, the device collects the user's voice and facial expression data. In this step, the inputs are data from sensors and cameras, and the user's voice and facial expression data, and the output is raw data that is temporarily stored.

[0704] Step 2:

[0705] The terminal sends the collected data to the server at regular intervals. Communication is performed using a secure protocol (e.g., TLS). In this step, there is temporarily stored data as input, and data is generated to be sent to the server as output.

[0706] Step 3:

[0707] The server converts the received data into a parseable format. It performs data cleaning, checking for outliers and missing values, and formatting the data. In this step, the raw data received is the input, and the cleaned data is produced as the output.

[0708] Step 4:

[0709] The server inputs the formatted data into a machine learning model to analyze the pet's behavior patterns. An anomaly detection model is applied to detect abnormal behavior. In this step, the input is cleaned data, and the output is the anomaly detection result.

[0710] Step 5:

[0711] The server analyzes the user's voice and facial expression data and applies an emotion recognition algorithm to determine whether the user is stressed or relaxed. In this step, the user's voice and facial expression data are the input, and the emotion recognition result is generated as the output.

[0712] Step 6:

[0713] The server integrates anomaly detection results and sentiment identification results to generate appropriate notification content and countermeasures. It adjusts the text and wording of the notification based on the user's emotional state. In this step, the input is anomaly detection results and sentiment identification results, and the output is the adjusted notification content.

[0714] Step 7:

[0715] The terminal receives notifications from the server and displays them through the user interface. The user reviews the notification content and takes appropriate action. In this step, the input is the adjusted notification content, and the output is the notification sent to the user.

[0716] Step 8:

[0717] The user checks the pet's condition and takes necessary actions based on the notification. Feedback on observations and actions taken is provided through the application. In this step, the user's observation and action information is the input, and feedback information is generated as the output.

[0718] Step 9:

[0719] The server receives feedback from the user and uses it to retrain the machine learning model and sentiment recognition algorithm. In this step, the received feedback information is taken as input, and the updated model is produced as output.

[0720] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0721] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0722] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0723] [Third Embodiment]

[0724] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0725] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0726] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0727] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0728] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0729] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0730] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0731] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0732] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0733] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0734] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0735] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0736] This invention is a system designed to detect the health status and illnesses of pets at an early stage. This system collects behavioral data from pets using multiple sensors and cameras, and analyzes this data on a server to detect abnormalities. The specific configuration and operation of this system are described in detail below.

[0737] Data collection

[0738] The device collects behavioral data in real time via sensors and cameras attached to the pet. For example, it tracks the pet's movement patterns with a GPS sensor and records its activity level with an accelerometer. It captures the pet's behavior as video using a camera, and records the pet's movements as video data. This data is temporarily stored in a buffer within the device.

[0739] Data transmission

[0740] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security. The data is time-stamped to maintain its order precisely.

[0741] Data analysis and anomaly detection

[0742] The server analyzes the received data. First, it converts the data into a format that can be analyzed, and then cleans it by checking for outliers and missing values. Next, it extracts features to compare with past behavioral data. Specific features include exercise levels, meal frequency, and toilet usage frequency. Based on these features, a machine learning model is used to detect anomalies. When an abnormal behavioral pattern is detected, information indicating that the behavior is abnormal is generated.

[0743] Anomaly notification and countermeasures provided

[0744] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the detected abnormal behavior. The terminal displays this information to the user (pet owner) through a user interface and provides appropriate countermeasures. For example, if an anomaly is detected, an alert such as "Your pet is behaving unusually. Please check on them" is displayed to the user.

[0745] User behavior and feedback

[0746] Based on the notified abnormal information, the user checks the pet's current condition and takes appropriate action as needed. The user observes the pet's condition and takes action, such as taking it to a veterinarian. Afterwards, they provide feedback through the application regarding the actual observation results and the actions taken.

[0747] Gathering feedback and retraining the model

[0748] The device sends feedback information provided by the user to the server. The server then uses this feedback information to retrain the anomaly detection model, improving the system's accuracy. This allows for more accurate monitoring of the pet's health.

[0749] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. The device then sends a notification to the user stating, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian." Based on this information, the user can check their pet's paws and, if there is a problem, take it to a veterinarian, enabling early detection and treatment.

[0750] The following describes the processing flow.

[0751] Step 1:

[0752] The device collects behavioral data in real time using sensors and cameras attached to the pet. Specifically, it tracks movement patterns with a GPS sensor and records activity levels with an accelerometer. At the same time, it captures behavior as video with a camera and detects specific actions (e.g., walking, sitting, licking).

[0753] Step 2:

[0754] The device temporarily stores the collected data in a buffer. The buffer is time-stamped to accurately preserve the order of the data. For example, movement data, activity level data, and video data are stored in chronological order.

[0755] Step 3:

[0756] The terminal sends data from a buffer to the server at regular intervals. This transmission process uses secure protocols such as TLS to ensure data consistency and security. For example, data batches are sent to the server every minute.

[0757] Step 4:

[0758] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and normalizes the sensor's numerical data. For example, it converts the video data into JPEG frames and scales the acceleration data within a specified range.

[0759] Step 5:

[0760] The server performs data cleaning. It checks for outliers (e.g., obviously incorrect numbers or corrupted data) and missing values, and removes or fills in the invalid data. For example, it might fill in abnormal spikes in activity level data or missing consecutive frames.

[0761] Step 6:

[0762] The server extracts features to compare with past pet behavior data stored in the database. For example, it calculates features such as daily exercise, number of meals, and frequency of litter box use.

[0763] Step 7:

[0764] The server uses a machine learning model to analyze features and detect anomalies. Specifically, it compares abnormal behavior (e.g., continuous licking of paws, sudden decrease in activity level) with normal behavior patterns to detect abnormal behavior. The model is trained based on past data and feedback.

[0765] Step 8:

[0766] The server notifies the terminal of the detected anomaly. The anomaly information includes details such as the type of abnormal behavior detected, the date and time, and the frequency.

[0767] Step 9:

[0768] The device notifies the user of any abnormal information it receives via the user interface. For example, a smartphone application might display a pop-up notification with a message such as, "Your pet is behaving unusually. Please check on it."

[0769] Step 10:

[0770] The user checks the notification, observes their pet's condition, and takes necessary action. For example, they might check their pet's paws and consult a veterinarian if any abnormalities are found.

[0771] Step 11:

[0772] Users provide feedback through the application, including observations and actions taken. Specifically, they input text information about their pet's current condition, as well as veterinary diagnoses.

[0773] Step 12:

[0774] The device sends user feedback information to the server. The transmitted feedback includes specific observations and details of treatment.

[0775] Step 13:

[0776] The server analyzes the received feedback information and retrains the machine learning model based on the new insights. This improves the system's anomaly detection accuracy and provides more precise diagnoses.

[0777] (Example 1)

[0778] Next, we will describe Example 1. 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."

[0779] In modern society, it is extremely important to monitor pets' health in real time and detect abnormalities early. However, with conventional systems, pet owners often fail to notice their pets' illnesses until it is too late, making early treatment difficult. Furthermore, there is a lack of efficient systems for collecting and analyzing pet behavioral data and developing appropriate countermeasures based on that data.

[0780] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0781] In this invention, the server includes means for collecting pet behavior data using a detection device, means for temporarily storing the collected behavior data, means for transmitting the stored behavior data to an information processing device, means for analyzing the transmitted behavior data and applying a machine learning algorithm to detect anomalies, means for notifying a display terminal of the detected anomaly information, means for displaying the notified anomaly information and providing the user with appropriate countermeasures, means for collecting user feedback and transmitting it to the information processing device, and means for retraining the machine learning algorithm using the transmitted feedback. This enables real-time monitoring of the pet's health status, early detection of anomalies, and prompt and appropriate responses.

[0782] A "detection device" is a device that includes sensors and cameras used to collect pet behavioral data.

[0783] A "location detection sensor" is a sensor used to obtain a pet's location information, and specifically includes GPS sensors.

[0784] A "activity detection sensor" is a sensor used to measure a pet's activity level, and specifically includes accelerometers and similar sensors.

[0785] "Behavioral data" refers to data including the pet's location, movements, and activity level.

[0786] "Means of temporary storage" refers to memory or storage for temporarily storing collected data.

[0787] An "information processing device" is a computing device that analyzes received data and outputs the results.

[0788] A "machine learning algorithm" is a mathematical model used to detect anomalies based on past data.

[0789] A "display terminal" is a device used to notify users of abnormal information transmitted from a server. Specifically, this includes smartphones and tablets.

[0790] "Users" refer to pet owners or caretakers.

[0791] "Feedback" refers to information about what users observed and the actions they took.

[0792] This invention relates to a system for early detection of the health status or illness of pets. This system uses multiple hardware and software components to collect and analyze pet behavioral data and detect abnormalities. Specifically, it uses a combination of sensors and cameras as detection devices, an information processing device for analysis and notification, and a terminal for collecting feedback.

[0793] The device collects real-time behavioral data of the pet using location detection sensors (e.g., GPS sensors) and activity detection sensors (e.g., accelerometers) attached to the pet. This data is stored in the device's temporary storage. The collected data is analyzed and notifications are sent according to the following processing flow.

[0794] Hardware and software usage

[0795] 1. Detection device:

[0796] Location sensors collect information about a pet's current location. For example, a GPS sensor tracks the pet's movement patterns.

[0797] The activity detection sensor records the pet's activity level and uses an accelerometer to measure how active the pet is.

[0798] 2. Information processing device:

[0799] Data transmission: The terminal sends the collected data to the server (information processing device) at regular intervals (e.g., every 5 minutes). This process is carried out using a secure protocol (e.g., TLS).

[0800] Data Analysis and Anomaly Detection: The server analyzes the received data and detects anomalies. First, it converts the data into a parseable format (e.g., CSV or JSON). Next, it cleans the data and extracts features (such as exercise levels, meal frequency, and toilet usage frequency). Based on these features, it applies machine learning algorithms to detect abnormal behavior.

[0801] 3. Notice:

[0802] Anomaly Notification: The server notifies the terminal of any detected anomalies. The terminal displays this information to the user through a user interface (such as a smartphone app). For example, an alert might be displayed stating, "Your pet is behaving unusually. Please check on it."

[0803] 4. Gathering feedback:

[0804] User Behavior and Feedback: Users check the status of their pets based on the notified abnormal information and take appropriate measures as needed. Users provide feedback on their observations and actions taken through the application.

[0805] Model retraining: The terminal sends the provided feedback information to the server, which then retrains the anomaly detection model based on this feedback. This improves the system's accuracy and enables more precise anomaly detection in the future.

[0806] Specific example

[0807] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws, and this video data is sent from the device to a server. The server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. As a result, the device sends a notification to the user stating, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian." Based on this information, the user can check their pet's paws and, if there is a problem, take it to a veterinarian, enabling early detection and treatment.

[0808] An example of a prompt statement is: "Use a system that collects pet behavior data using cameras and sensors and analyzes it on a server to monitor the pet's health in real time and detect abnormalities."

[0809] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0810] Step 1:

[0811] Data collection

[0812] The device collects behavioral data in real time using location sensors (e.g., GPS sensors) and activity sensors (e.g., accelerometers) attached to the pet. Inputs include signals from the sensors (location information, acceleration data). These sensor signals are converted into digital data and stored in a temporary buffer memory within the device. Outputs consist of a set of location information and activity data.

[0813] Specific actions:

[0814] The GPS sensor collects location information every second.

[0815] The accelerometer records the pet's movements every 0.1 seconds.

[0816] The collected data is temporarily stored in a buffer.

[0817] Step 2:

[0818] Data transmission

[0819] The terminal sends data stored in a buffer to the server at regular intervals (e.g., every 5 minutes). The input is location and momentum data stored in the buffer, and the output is data packets sent to the server. Secure protocols such as TLS are used for transmission, and the order of the data is maintained by timestamping each data packet.

[0820] Specific actions:

[0821] The data stored in the buffer is read and sent to the server using the TLS protocol.

[0822] After transmission is complete, the buffer is cleared to prepare for the next data collection.

[0823] Step 3:

[0824] Data Analysis

[0825] The server receives and analyzes data sent from the terminal. The input is location and activity data sent from the terminal, and the output is analyzed features (e.g., activity level, number of meals, frequency of toilet use, etc.) and anomaly detection results. First, the raw data is converted into an analyzable format (e.g., CSV or JSON), and outliers and missing values ​​are checked and cleaned. Then, specific features are extracted, and abnormal behavior is analyzed using a machine learning algorithm.

[0826] Specific actions:

[0827] Convert the raw data to a different format and clean the data as needed.

[0828] Extract features and organize them into an analyzable format.

[0829] Machine learning algorithms are applied to detect anomalies.

[0830] Step 4:

[0831] Abnormal notification

[0832] The server notifies the terminal of any anomalies detected. The input consists of the analyzed data and the results of the anomaly detection, while the output is the notification message sent to the terminal. The notification includes the type, date, time, and frequency of the detected anomaly.

[0833] Specific actions:

[0834] A notification message is constructed based on the detected anomaly data.

[0835] Send a message to the device and display it on the user interface.

[0836] Step 5:

[0837] Feedback Collection

[0838] The device collects user feedback and sends it to the server. The input is feedback information provided by the user, and the output is feedback data sent to the server. The feedback includes detailed information such as the user's observations of the pet's condition, the actions taken, and the results.

[0839] Specific actions:

[0840] Collect feedback through the user interface.

[0841] The collected feedback is sent to the server.

[0842] Step 6:

[0843] Model Retraining

[0844] The server retrains its machine learning algorithm based on feedback received from the user. The input consists of feedback data and historical behavioral data, and the output is an updated anomaly detection model. This improves the accuracy of anomaly detection.

[0845] Specific actions:

[0846] Integrate feedback data with past behavioral data.

[0847] Retrain the anomaly detection model and apply the new model.

[0848] (Application Example 1)

[0849] Next, we will explain Application Example 1. In the following explanation, 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."

[0850] Properly monitoring a pet's health while it is traveling and detecting abnormalities early is crucial for ensuring their safety and well-being. Conventional pet health monitoring systems are typically used within the home or in specific locations, and monitoring during travel has not been considered. As a result, it has been difficult to ascertain a pet's health status in a timely manner, especially during long journeys. Furthermore, while a quick and accurate response is required when an abnormality is detected in a pet, there is a challenge in that appropriate means of response are not readily available while the pet is traveling.

[0851] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0852] In this invention, the server includes means for collecting pet behavior data using sensors and cameras, means for temporarily storing the collected behavior data, means for transmitting the stored behavior data to the server, means for analyzing the transmitted behavior data and applying a machine learning model to detect anomalies, means for notifying the terminal of the detected anomaly, means for displaying the notified anomaly information and providing the user with appropriate countermeasures, means for collecting user feedback and transmitting it to the server, means for retraining the machine learning model using the transmitted feedback, means for collecting pet behavior data using sensors and cameras mounted in the autonomous vehicle, means for processing the data in real time and detecting anomalies in cooperation with the system in the autonomous vehicle, and means for notifying the vehicle's dashboard display and smartphone when an anomaly is detected. This makes it possible to appropriately monitor the health status of pets even while they are in transit and to detect anomalies early.

[0853] "Pet behavioral data" refers to information related to a pet's movement patterns, activity level, and health status.

[0854] A "sensor" is a device that captures environmental information or physical quantities and converts them into electrical signals, etc. Specific examples include GPS sensors and accelerometers.

[0855] A "camera" is a device that captures images and videos, and includes 360-degree cameras and night vision cameras.

[0856] "Means of temporary storage" refers to memory or buffers used to retain data for a certain period of time.

[0857] "Means of sending to the server" refers to a communication module that has the function of transferring collected data to a computer in a remote location.

[0858] A "machine learning model" refers to an algorithm that uses past data to recognize patterns and then analyzes and predicts new data.

[0859] "Means for detecting anomalies" refers to a function that uses machine learning models to identify anomalies that deviate from normal behavioral patterns.

[0860] "Means of notifying the terminal" refers to communication devices or applications that have the function of transmitting anomaly detection information to the user.

[0861] A "dashboard display" refers to a display device installed inside an autonomous vehicle, used to visually display notification information and warnings.

[0862] A "smartphone" refers to a mobile device equipped with multiple functions, such as voice communication, data communication, and application execution.

[0863] "User feedback" refers to responses and information provided by users regarding the system's operation and notifications.

[0864] "Methods for processing data in real time" refer to functions that perform analysis and decision-making almost simultaneously with the moment data is collected.

[0865] An "autonomous vehicle" refers to a vehicle that operates autonomously without the intervention of a driver.

[0866] This invention relates to a system designed for the early detection of pet health issues and illnesses. This system uses multiple sensors and cameras to collect behavioral data for monitoring pet health within autonomous vehicles, and analyzes this data on a server to detect abnormalities. The specific configuration and operation of this system are described in detail below.

[0867] Data collection

[0868] The terminal (a platform installed inside the autonomous vehicle) collects behavioral data in real time via sensors and cameras attached to the pet. For example, a GPS sensor tracks the pet's movement patterns, and an accelerometer records the pet's activity level. Cameras installed inside the vehicle capture the pet's actions as video, and the pet's movements are captured as video data. This data is temporarily stored in a buffer within the terminal. Specifically, if the pet is frequently moving or active inside the vehicle, its movements are recorded with a 360-degree camera.

[0869] Data transmission

[0870] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (TLS) to ensure data consistency and security. The data is time-stamped to accurately preserve its order.

[0871] Data analysis and anomaly detection

[0872] The server analyzes the received data. First, it converts the data into a format that can be analyzed, and then cleans it by checking for outliers and missing values. Next, it extracts features to compare with past behavioral data. Specific features include exercise levels, meal frequency, and toilet usage frequency. Based on these features, a generative AI model is used to detect anomalies. When an abnormal behavioral pattern is detected, information that has been judged as abnormal is generated.

[0873] Anomaly notification and countermeasures provided

[0874] The server notifies the terminal of any detected anomalies. The terminal provides an interface to notify the user (pet owner) of this information, displaying it via voice, dashboard display, smartphone, etc. For example, if an anomaly is detected, an alert such as "Your pet is behaving unusually. Please check on them" will be displayed on the vehicle's dashboard display and smartphone.

[0875] User behavior and feedback

[0876] Based on the notified abnormal information, users can check their pet's condition and take appropriate action. For example, if the camera captures the pet frequently licking its front paws, the user can use that information to check the pet's paws and take it to the veterinarian if any abnormalities are found. Users can also provide feedback on their observations and the actions taken through the application.

[0877] Gathering feedback and retraining the model

[0878] The device sends feedback information provided by the user to the server. The server then retrains the anomaly detection model based on this feedback information, improving the system's accuracy. This makes it possible to monitor the pet's health more accurately.

[0879] As a concrete example, by using a prompt statement such as, "Write a program that analyzes sensor and camera data in real time to monitor the health of pets and detect abnormalities," the system's operation can be made to work exactly as specified.

[0880] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0881] Step 1:

[0882] The device collects behavioral data in real time using sensors and cameras attached to the pet. The input is raw data from the sensors and camera, and the output is data temporarily stored in a buffer within the device. Specifically, it simultaneously acquires the pet's location information from the GPS sensor, activity level from the accelerometer, and video data from the camera, and records each of these in the buffer.

[0883] Step 2:

[0884] The terminal sends data stored in a buffer to the server at regular intervals. The input here is the behavioral data stored in the buffer, and the output is the data sent to the server. The TLS protocol is used for this transmission, ensuring data consistency and security. Specifically, once data has accumulated in the buffer for a certain period, it is uploaded to the server in a single batch.

[0885] Step 3:

[0886] The server analyzes the received data. The input is behavioral data sent from the terminal, and the output is data converted into a parseable format. Specifically, after parsing the data, it checks for outliers and missing values ​​and performs cleaning. It also analyzes timestamp information to maintain the integrity of time-series data.

[0887] Step 4:

[0888] The server extracts features from the analyzed data and uses them to apply a generative AI model to detect anomalies. The input in this process is cleaned data, and the output is the anomaly detection result. Specifically, it extracts features such as exercise volume, meal frequency, and toilet usage frequency, and inputs these into the generative AI model to detect anomaly patterns.

[0889] Step 5:

[0890] The server notifies the terminal of any detected anomalies. The input is the result of the anomaly detection, and the output is the data for notification. Specifically, as soon as an anomaly is detected, the server sends that information to the terminal in real time and prepares to issue a warning to the user.

[0891] Step 6:

[0892] The terminal notifies the user of the received anomaly information. The input here is the anomaly information sent from the server, and the output is the notification to the user. Specifically, it displays the anomaly details and corrective actions using the vehicle's dashboard display or the smartphone's notification function.

[0893] Step 7:

[0894] The user checks the pet's condition based on the notified abnormal information and takes appropriate action as needed. The input is notifications from the terminal, and the output is user feedback. Specifically, the user actually observes the pet's condition and reports the results to the system via the terminal.

[0895] Step 8:

[0896] The terminal sends user-provided feedback information to the server. The input is the user's feedback, and the output is the data sent to the server. Specifically, it collects feedback information and sends it to the server along with a timestamp.

[0897] Step 9:

[0898] The server retrains the machine learning model using the feedback it receives. The input is the feedback data sent by the user, and the output is the new machine learning model with improved accuracy. Specifically, it incorporates the collected feedback data as part of the training data and updates the generative AI model.

[0899] By following the steps outlined above, it is possible to effectively implement a system that monitors a pet's health in real time and detects abnormalities early.

[0900] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0901] This invention is a system for supporting pet health management, which collects and analyzes pet behavioral data in real time and detects abnormalities. Furthermore, the invention incorporates an emotion engine that recognizes the user's emotions, and adds a function to adjust notifications and countermeasures based on the user's emotions. The specific configuration and embodiments of this system are described below.

[0902] Data collection

[0903] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. The GPS sensor tracks the pet's movement patterns, and the accelerometer records the pet's activity level. Simultaneously, a camera is used to capture the pet's behavior as video and detect specific actions (e.g., walking, sitting, licking). This data is temporarily stored in a buffer within the device.

[0904] Data transmission and analysis

[0905] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security.

[0906] The server converts the received data into an analyzable format, checks for outliers and missing values, and performs data cleaning. Next, it extracts features for comparison with past pet behavior data and uses a machine learning model to detect anomalies.

[0907] Using an Emotion Engine

[0908] This system includes an emotion engine that recognizes the user's emotions. The terminal captures the user's voice and facial expressions and sends that data to the server.

[0909] The server analyzes this data and applies emotion recognition algorithms. For example, it identifies whether the user is stressed or relaxed. This analysis result is taken into consideration when notifying users of abnormal information about their pet.

[0910] Anomaly notification and countermeasures provided

[0911] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the pet's abnormal behavior. The content and method of notification are adjusted based on the analysis results of the emotion engine. For example, if the user is stressed, the notification will be expressed in a softer tone, while if they are relaxed, more detailed information will be provided.

[0912] The device displays this information to the user through a user interface and provides appropriate countermeasures. For example, a message such as, "Your pet is behaving unusually. Please check on it," might be displayed.

[0913] User behavior and feedback

[0914] Users check notifications, observe their pet's condition, and take necessary actions. For example, they might check their pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken.

[0915] Gathering feedback and retraining the model

[0916] The device sends user-provided feedback information and sentiment data to the server.

[0917] Based on this feedback, the server retrains its anomaly detection model and emotion recognition algorithm to improve the system's accuracy and personalization. This allows for more accurate monitoring of pet health and the provision of optimal notifications and actions tailored to the user's emotional state.

[0918] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. Simultaneously, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and don't panic." The user can then use this information to check on their pet's condition and, if necessary, take it to a veterinarian, enabling early detection and treatment.

[0919] The following describes the processing flow.

[0920] Step 1:

[0921] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. Specifically, the GPS sensor periodically records the pet's location coordinates, and the accelerometer monitors the pet's activity level (e.g., intensity and frequency of movement). Furthermore, a camera is used to capture the pet's behavior as video, and specific actions (e.g., walking, sitting, licking) are automatically detected.

[0922] Step 2:

[0923] The device temporarily stores the collected behavioral data in a buffer. The buffer is time-stamped to accurately preserve the order of the data. For example, daily movement data, activity level data, and video data are stored in chronological order.

[0924] Step 3:

[0925] The terminal sends data stored in a buffer to the server at regular intervals (e.g., every minute). Data communication is securely performed using the TLS (Transport Layer Security) protocol, ensuring data consistency and security.

[0926] Step 4:

[0927] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and normalizes the sensor's numerical data. For example, the video data is converted into JPEG frames, and the acceleration data is scaled within a specified range.

[0928] Step 5:

[0929] The server performs data cleaning. It checks for outliers and missing values ​​and removes or fills in invalid data. For example, it identifies and processes abnormal spikes in activity level data or missing consecutive frames.

[0930] Step 6:

[0931] The server extracts features to compare with past pet behavior data stored in the database. For example, it calculates features such as daily exercise, number of meals, and frequency of litter box use.

[0932] Step 7:

[0933] The server uses a machine learning model to analyze features and detect abnormal pet behavior. For example, it can detect abnormal behavior such as "constantly licking its paws" by comparing it to normal behavior patterns. The machine learning model is trained based on past training data.

[0934] Step 8:

[0935] The device captures the user's voice and facial expressions and sends that data to a server. For example, it uses the smartphone's microphone and camera to record audio and video while the user is observing their pet.

[0936] Step 9:

[0937] The server analyzes the transmitted audio and video data and uses an emotion engine to recognize the user's emotions. For example, it can determine whether the user is stressed or relaxed based on changes in voice tone and facial expressions.

[0938] Step 10:

[0939] The server determines the content and method of notification based on the detected anomaly information and the results of the emotion engine's analysis. For example, if the user is feeling stressed, it will generate a gentle message such as, "Your pet is licking its paws repeatedly. We understand you are worried, but please take your time and check on it carefully."

[0940] Step 11:

[0941] The device displays the generated notification content to the user and suggests appropriate actions. For example, it might display a message via the smartphone's notification function saying, "Your pet is behaving unusually. We recommend consulting a veterinarian."

[0942] Step 12:

[0943] The user checks the notification, observes the pet's condition, and takes appropriate action (e.g., takes it to the vet). They then input feedback about the observation results and the actions taken into the application.

[0944] Step 13:

[0945] The device sends user-provided feedback and sentiment data to the server. For example, it sends text information entered by the user into the application and diagnostic results.

[0946] Step 14:

[0947] The server uses feedback information and sentiment data to retrain machine learning models and sentiment recognition algorithms, improving the system's accuracy. This allows the system to more accurately monitor pet health and provide optimal notifications and actions tailored to the user's emotional state.

[0948] (Example 2)

[0949] Next, we will describe Example 2. 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."

[0950] A current challenge in pet health management is the difficulty in quickly and accurately detecting abnormal behavior in pets and notifying users of this information in an appropriate manner. Furthermore, because the user's emotional state is not taken into consideration, the content of the notifications may cause excessive stress to the user. To solve these problems, this invention aims to provide a system that collects and analyzes not only pet behavior data but also user emotional data, and provides optimal notifications and countermeasures tailored to individual situations.

[0951] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0952] In this invention, the server includes means for analyzing pet behavior data and applying a machine learning model to detect anomalies, means for analyzing user emotion data and applying an emotion recognition algorithm, and means for adjusting anomaly notifications based on the user's emotional state. This enables high-precision detection of pet behavioral anomalies and the provision of appropriate notifications and countermeasures that take into account the user's emotional state.

[0953] "Pet behavioral data" refers to data that shows a pet's movement patterns, activity levels, and specific behaviors (e.g., walking, sitting, licking).

[0954] A "sensor" is a device used to measure the state of the environment or objects. Specifically, this includes GPS sensors and accelerometers.

[0955] A "camera" is a device used to capture video data.

[0956] A "buffer" is a memory area used to temporarily store data.

[0957] A "server" is a computer system that processes data over a network.

[0958] A "machine learning model" is a model built using algorithms to make predictions and classifications based on data.

[0959] "Emotional data" refers to data that indicates the emotional state of a user, obtained from their voice and facial expressions.

[0960] An "emotion identification algorithm" is an algorithm that analyzes emotional data to identify a user's emotional state (e.g., stress, relaxation).

[0961] An "abnormality notification" is a notification sent to the user when an abnormality is detected in the pet's behavior.

[0962] "User feedback" refers to information about observations and actions taken by users that they provide to the system.

[0963] "Retraining" is the process of improving a machine learning model using newly collected data.

[0964] This invention is a system for supporting pet health management, specifically by collecting and analyzing pet behavioral data in real time and detecting abnormalities. Furthermore, the invention incorporates an emotion engine that recognizes the user's emotions, and adds a function to adjust notifications and countermeasures based on the user's emotions.

[0965] Data collection

[0966] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. The GPS sensor tracks the pet's movement patterns, and the accelerometer records the pet's activity level. Simultaneously, a camera is used to capture the pet's behavior as video and detect specific actions (e.g., walking, sitting, licking). This data is temporarily stored in a buffer within the device.

[0967] Data transmission and analysis

[0968] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security.

[0969] The server converts the received data into an analyzable format, checks for outliers and missing values, and performs data cleaning. Next, it extracts features for comparison with past pet behavior data and uses a machine learning model to detect anomalies.

[0970] Using an Emotion Engine

[0971] This system includes an emotion engine that recognizes the user's emotions. The terminal captures the user's voice and facial expressions and sends that data to the server.

[0972] The server analyzes this data and applies emotion recognition algorithms. For example, it identifies whether the user is stressed or relaxed. This analysis is taken into consideration when notifying users of abnormal information about their pet.

[0973] Anomaly notification and countermeasures provided

[0974] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the pet's abnormal behavior. The content and method of notification are adjusted based on the analysis results of the emotion engine. For example, if the user is stressed, the notification will be expressed in a softer tone, while if they are relaxed, more detailed information will be provided.

[0975] The device displays this information to the user through a user interface and provides appropriate countermeasures. For example, a message such as, "Your pet is behaving unusually. Please check on it," might be displayed.

[0976] User behavior and feedback

[0977] Users check notifications, observe their pet's condition, and take necessary actions. For example, they might check their pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken.

[0978] Gathering feedback and retraining the model

[0979] The device sends user-provided feedback information and sentiment data to the server.

[0980] Based on this feedback, the server retrains its anomaly detection model and emotion recognition algorithm to improve the system's accuracy and personalization. This allows for more accurate monitoring of pet health and the provision of optimal notifications and actions tailored to the user's emotional state.

[0981] Specific example

[0982] For example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. Simultaneously, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and don't panic." The user can then use this information to check on their pet's condition and, if necessary, take it to a veterinarian, enabling early detection and treatment.

[0983] Example of a prompt:

[0984] "Please explain how the system works to detect abnormal pet behavior and provide appropriate notifications and countermeasures based on the user's emotions."

[0985] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0986] Step 1:

[0987] Data collection

[0988] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. First, it obtains the pet's current location from the GPS sensor every second, and measures the acceleration along the X, Y, and Z axes in milliseconds using the accelerometer. Furthermore, it captures the pet's behavior as video at 30 frames per second using a camera, and uses an image recognition algorithm to identify specific actions. This data is temporarily stored in a buffer. Sensor data and video data are collected as input and stored in the buffer as output.

[0989] Step 2:

[0990] Data transmission

[0991] The device sends activity data stored in a buffer to the server every 30 seconds. The transmitted data includes GPS data, acceleration data, and video data. The transmission process uses the TLS protocol to ensure data security and consistency. It takes data from the buffer as input and sends it to the server as output.

[0992] Step 3:

[0993] Data Analysis

[0994] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and identifies the pet's movements using an image recognition algorithm. GPS data and acceleration data are synchronized using timestamps, and outliers and missing values ​​are filtered out. Features are extracted using the cleaned data. The server takes the received data as input and obtains the analyzed data and features as output.

[0995] Step 4:

[0996] Collection and transmission of emotional data

[0997] The device uses the front camera and microphone to collect user emotion data. The camera captures the user's facial expressions, and the microphone records their voice. This data is temporarily stored in a buffer and sent to the server at regular intervals. User facial expression data and voice data are taken as input, stored in a buffer as output, and sent to the server.

[0998] Step 5:

[0999] Emotion analysis

[1000] The server analyzes the received emotional data. It uses a facial recognition algorithm to analyze the user's facial features and identify emotions. It also uses a speech recognition algorithm to analyze audio data and estimate emotions from voice tone and speed. Using these analysis results, it determines whether the user is stressed or relaxed. It takes the received emotional data as input and obtains the emotional state as output.

[1001] Step 6:

[1002] Anomaly detection

[1003] The server uses a machine learning model to detect anomalies based on features. For example, if a pet repeatedly performs the same action within a specific range (e.g., continuously licking its front paws), this is detected as an anomaly. The server classifies the type of anomaly and associates it with specific behavioral patterns. It takes cleaned data and features as input and obtains anomaly detection results as output.

[1004] Step 7:

[1005] Abnormal notification

[1006] The server generates a notification message based on the anomaly detection results. This message includes the type of abnormal behavior, date and time, frequency, and content based on the sentiment analysis results. It provides gentle information if the user is stressed, and detailed information if they are relaxed. The terminal displays the received notification to the user. For example, it displays the notification content using smartphone push notifications or the interface of a dedicated app. It takes anomaly detection results and sentiment analysis results as input, and displays the generated notification to the user as output.

[1007] Step 8:

[1008] Collecting user feedback

[1009] The user checks the notification and actually observes the pet's condition. For example, they check the pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken. The application takes the notification content as input and provides information regarding observations and actions taken as output.

[1010] Step 9:

[1011] Feedback analysis and model retraining

[1012] The device sends user-provided feedback and sentiment data to the server. The server uses this feedback to retrain its anomaly detection model and sentiment recognition algorithm. Specifically, it uses the feedback information as labeled data to improve the accuracy and personalization of the machine learning model. It retrieves the feedback information sent as input and obtains the improved anomaly detection model and sentiment recognition algorithm as output.

[1013] (Application Example 2)

[1014] Next, we will explain application example 2. In the following explanation, 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."

[1015] Conventional pet health management systems have the functionality to collect and analyze pet behavior data in real time, but they do not provide notifications or countermeasures that take into account the user's emotional state. As a result, when the user is stressed, the notification content may be inappropriate, and there may be cases where an abnormality in the pet cannot be addressed quickly. The present invention aims to provide a more effective pet health management system that takes the user's emotions into consideration.

[1016] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting and transmitting user emotion data to the server, means for analyzing the transmitted emotion data and applying an emotion identification algorithm, and means for adjusting countermeasures based on the detected anomaly information and the user's emotional state and notifying the terminal. This makes it possible to monitor the pet's health status in real time while providing appropriate notifications and countermeasures according to the user's emotional state.

[1017] "Behavioral data" refers to data that shows a pet's movement patterns and activity levels, and is collected using GPS sensors and accelerometers.

[1018] A "sensor" is a device that detects physical information about the environment or objects, such as a GPS sensor or an accelerometer, and outputs that information as data.

[1019] A "camera" is a device used to capture a pet's behavior as video, and it is equipped with the function to detect specific actions.

[1020] A "server" refers to a computer system used to analyze collected data, detect anomalies, and analyze user emotions.

[1021] A "machine learning model" includes data analysis algorithms used to detect anomalies based on pet behavior data.

[1022] "Emotion recognition" refers to an algorithm that identifies a user's emotional state by analyzing their voice and facial expression data.

[1023] An "abnormal" behavior refers to a state in which a pet exhibits behavior that is clearly different from its normal behavioral patterns, suggesting a potential health problem.

[1024] "Notifications" refer to information sent to inform users about detected anomalies or measures taken based on the user's emotional state.

[1025] "Countermeasures" include instructions that show how to respond to detected anomalies or what actions to take.

[1026] "User feedback" refers to information provided by users regarding their pet's condition and the measures taken, which is used to improve the system's performance.

[1027] "Retraining" refers to the process of improving the performance of machine learning models and sentiment recognition algorithms using newly collected data.

[1028] This invention is a system for supporting pet health management, which uses sensors and cameras to collect and analyze pet behavior data in real time. Furthermore, by collecting and analyzing user emotional data, the system can provide notifications and countermeasures based on the user's emotional state. Based on this embodiment, a method for effectively managing a pet's health will be described.

[1029] System Configuration

[1030] The system mainly consists of the following components:

[1031] Sensors: These are GPS sensors and accelerometers attached to the pet, which are used to obtain the pet's location information and movement data.

[1032] Camera: A high-resolution camera that captures specific actions of your pet (e.g., walking, sitting, licking).

[1033] Devices (smartphones, tablets, etc.): Temporarily store collected data and send it to the server as needed.

[1034] Server: A computer system that performs data analysis, analyzing collected data using machine learning models and sentiment recognition algorithms.

[1035] User interface: An application used to notify users of abnormal information and countermeasures.

[1036] System operation

[1037] 1. Data Collection: Pet behavior data collected by sensors (e.g., GPS data, accelerometer data) and behavioral video data captured by cameras are transmitted to the device. The device also collects user voice and facial expression data.

[1038] 2. Data transmission: The collected data is sent to the server at regular intervals using a secure protocol.

[1039] 3. Analysis and Anomaly Detection: The server converts the received data into an analyzable format and uses machine learning models to detect abnormal pet behavior. Furthermore, it analyzes the user's emotional state using an emotion recognition algorithm.

[1040] 4. Provision of Notifications and Countermeasures: If an anomaly is detected, the server sends the most appropriate notification and countermeasure to the terminal based on the anomaly information and the user's emotional state. The terminal then notifies the user of this through the user interface.

[1041] 5. User Response and Feedback: Users will review notifications and take appropriate action regarding their pet's abnormal behavior. They will provide feedback on their observations and the measures taken through the application.

[1042] 6. System Improvement: The server retrains machine learning models and sentiment recognition algorithms based on user feedback to improve the system's accuracy and personalization.

[1043] Specific example

[1044] For example, if the camera captures a pet frequently licking its front paws, the video data is sent from the device to the server. The server analyzes the data and determines that the "continuous paw licking" behavior is abnormal. At the same time, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and do not panic." This allows the user to take appropriate action and effectively manage their pet's health.

[1045] Example of a prompt

[1046] "What should I do if my pet is frequently exhibiting signs of illness?"

[1047] This system allows for real-time monitoring of pet health and prompt, appropriate action. Furthermore, by providing notifications and solutions tailored to the user's emotional state, it reduces the user's mental burden, enabling more confident pet health management.

[1048] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1049] Step 1:

[1050] The device receives behavioral data collected by GPS and accelerometer sensors attached to the pet. It also collects video data of the pet's behavior captured by a camera. Simultaneously, the device collects the user's voice and facial expression data. In this step, the inputs are data from sensors and cameras, and the user's voice and facial expression data, and the output is raw data that is temporarily stored.

[1051] Step 2:

[1052] The terminal sends the collected data to the server at regular intervals. Communication is performed using a secure protocol (e.g., TLS). In this step, there is temporarily stored data as input, and data is generated to be sent to the server as output.

[1053] Step 3:

[1054] The server converts the received data into a parseable format. It performs data cleaning, checking for outliers and missing values, and formatting the data. In this step, the raw data received is the input, and the cleaned data is produced as the output.

[1055] Step 4:

[1056] The server inputs the formatted data into a machine learning model to analyze the pet's behavior patterns. An anomaly detection model is applied to detect abnormal behavior. In this step, the input is cleaned data, and the output is the anomaly detection result.

[1057] Step 5:

[1058] The server analyzes the user's voice and facial expression data and applies an emotion recognition algorithm to determine whether the user is stressed or relaxed. In this step, the user's voice and facial expression data are the input, and the emotion recognition result is generated as the output.

[1059] Step 6:

[1060] The server integrates anomaly detection results and sentiment identification results to generate appropriate notification content and countermeasures. It adjusts the text and wording of the notification based on the user's emotional state. In this step, the input is anomaly detection results and sentiment identification results, and the output is the adjusted notification content.

[1061] Step 7:

[1062] The terminal receives notifications from the server and displays them through the user interface. The user reviews the notification content and takes appropriate action. In this step, the input is the adjusted notification content, and the output is the notification sent to the user.

[1063] Step 8:

[1064] The user checks the pet's condition and takes necessary actions based on the notification. Feedback on observations and actions taken is provided through the application. In this step, the user's observation and action information is the input, and feedback information is generated as the output.

[1065] Step 9:

[1066] The server receives feedback from the user and uses it to retrain the machine learning model and sentiment recognition algorithm. In this step, the received feedback information is taken as input, and the updated model is produced as output.

[1067] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1068] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1069] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1070] [Fourth Embodiment]

[1071] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1072] As shown in Figure 7, the 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.

[1073] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1074] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1075] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1076] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1077] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1078] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1079] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1080] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[1081] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1082] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1083] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1084] This invention is a system designed to detect the health status and illnesses of pets at an early stage. This system collects behavioral data from pets using multiple sensors and cameras, and analyzes this data on a server to detect abnormalities. The specific configuration and operation of this system are described in detail below.

[1085] Data collection

[1086] The device collects behavioral data in real time via sensors and cameras attached to the pet. For example, it tracks the pet's movement patterns with a GPS sensor and records its activity level with an accelerometer. It captures the pet's behavior as video using a camera, and records the pet's movements as video data. This data is temporarily stored in a buffer within the device.

[1087] Data transmission

[1088] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security. The data is time-stamped to maintain its order precisely.

[1089] Data analysis and anomaly detection

[1090] The server analyzes the received data. First, it converts the data into a format that can be analyzed, and then cleans it by checking for outliers and missing values. Next, it extracts features to compare with past behavioral data. Specific features include exercise levels, meal frequency, and toilet usage frequency. Based on these features, a machine learning model is used to detect anomalies. When an abnormal behavioral pattern is detected, information indicating that the behavior is abnormal is generated.

[1091] Anomaly notification and countermeasures provided

[1092] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the detected abnormal behavior. The terminal displays this information to the user (pet owner) through a user interface and provides appropriate countermeasures. For example, if an anomaly is detected, an alert such as "Your pet is behaving unusually. Please check on them" is displayed to the user.

[1093] User behavior and feedback

[1094] Based on the notified abnormal information, the user checks the pet's current condition and takes appropriate action as needed. The user observes the pet's condition and takes action, such as taking it to a veterinarian. Afterwards, they provide feedback through the application regarding the actual observation results and the actions taken.

[1095] Gathering feedback and retraining the model

[1096] The device sends feedback information provided by the user to the server. The server then uses this feedback information to retrain the anomaly detection model, improving the system's accuracy. This allows for more accurate monitoring of the pet's health.

[1097] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. The device then sends a notification to the user stating, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian." Based on this information, the user can check their pet's paws and, if there is a problem, take it to a veterinarian, enabling early detection and treatment.

[1098] The following describes the processing flow.

[1099] Step 1:

[1100] The device collects behavioral data in real time using sensors and cameras attached to the pet. Specifically, it tracks movement patterns with a GPS sensor and records activity levels with an accelerometer. At the same time, it captures behavior as video with a camera and detects specific actions (e.g., walking, sitting, licking).

[1101] Step 2:

[1102] The device temporarily stores the collected data in a buffer. The buffer is time-stamped to accurately preserve the order of the data. For example, movement data, activity level data, and video data are stored in chronological order.

[1103] Step 3:

[1104] The terminal sends data from a buffer to the server at regular intervals. This transmission process uses secure protocols such as TLS to ensure data consistency and security. For example, data batches are sent to the server every minute.

[1105] Step 4:

[1106] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and normalizes the sensor's numerical data. For example, it converts the video data into JPEG frames and scales the acceleration data within a specified range.

[1107] Step 5:

[1108] The server performs data cleaning. It checks for outliers (e.g., obviously incorrect numbers or corrupted data) and missing values, and removes or fills in the invalid data. For example, it might fill in abnormal spikes in activity level data or missing consecutive frames.

[1109] Step 6:

[1110] The server extracts features to compare with past pet behavior data stored in the database. For example, it calculates features such as daily exercise, number of meals, and frequency of litter box use.

[1111] Step 7:

[1112] The server uses a machine learning model to analyze features and detect anomalies. Specifically, it compares abnormal behavior (e.g., continuous licking of paws, sudden decrease in activity level) with normal behavior patterns to detect abnormal behavior. The model is trained based on past data and feedback.

[1113] Step 8:

[1114] The server notifies the terminal of the detected anomaly. The anomaly information includes details such as the type of abnormal behavior detected, the date and time, and the frequency.

[1115] Step 9:

[1116] The device notifies the user of any abnormal information it receives via the user interface. For example, a smartphone application might display a pop-up notification with a message such as, "Your pet is behaving unusually. Please check on it."

[1117] Step 10:

[1118] The user checks the notification, observes their pet's condition, and takes necessary action. For example, they might check their pet's paws and consult a veterinarian if any abnormalities are found.

[1119] Step 11:

[1120] Users provide feedback through the application, including observations and actions taken. Specifically, they input text information about their pet's current condition, as well as veterinary diagnoses.

[1121] Step 12:

[1122] The device sends user feedback information to the server. The transmitted feedback includes specific observations and details of treatment.

[1123] Step 13:

[1124] The server analyzes the received feedback information and retrains the machine learning model based on the new insights. This improves the system's anomaly detection accuracy and provides more precise diagnoses.

[1125] (Example 1)

[1126] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1127] In modern society, it is extremely important to monitor pets' health in real time and detect abnormalities early. However, with conventional systems, pet owners often fail to notice their pets' illnesses until it is too late, making early treatment difficult. Furthermore, there is a lack of efficient systems for collecting and analyzing pet behavioral data and developing appropriate countermeasures based on that data.

[1128] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1129] In this invention, the server includes means for collecting pet behavior data using a detection device, means for temporarily storing the collected behavior data, means for transmitting the stored behavior data to an information processing device, means for analyzing the transmitted behavior data and applying a machine learning algorithm to detect anomalies, means for notifying a display terminal of the detected anomaly information, means for displaying the notified anomaly information and providing the user with appropriate countermeasures, means for collecting user feedback and transmitting it to the information processing device, and means for retraining the machine learning algorithm using the transmitted feedback. This enables real-time monitoring of the pet's health status, early detection of anomalies, and prompt and appropriate responses.

[1130] A "detection device" is a device that includes sensors and cameras used to collect pet behavioral data.

[1131] A "location detection sensor" is a sensor used to obtain a pet's location information, and specifically includes GPS sensors.

[1132] A "activity detection sensor" is a sensor used to measure a pet's activity level, and specifically includes accelerometers and similar sensors.

[1133] "Behavioral data" refers to data including the pet's location, movements, and activity level.

[1134] "Means of temporary storage" refers to memory or storage for temporarily storing collected data.

[1135] An "information processing device" is a computing device that analyzes received data and outputs the results.

[1136] A "machine learning algorithm" is a mathematical model used to detect anomalies based on past data.

[1137] A "display terminal" is a device used to notify users of abnormal information transmitted from a server. Specifically, this includes smartphones and tablets.

[1138] "Users" refer to pet owners or caretakers.

[1139] "Feedback" refers to information about what users observed and the actions they took.

[1140] This invention relates to a system for early detection of the health status or illness of pets. This system uses multiple hardware and software components to collect and analyze pet behavioral data and detect abnormalities. Specifically, it uses a combination of sensors and cameras as detection devices, an information processing device for analysis and notification, and a terminal for collecting feedback.

[1141] The device collects real-time behavioral data of the pet using location detection sensors (e.g., GPS sensors) and activity detection sensors (e.g., accelerometers) attached to the pet. This data is stored in the device's temporary storage. The collected data is analyzed and notifications are sent according to the following processing flow.

[1142] Hardware and software usage

[1143] 1. Detection device:

[1144] Location sensors collect information about a pet's current location. For example, a GPS sensor tracks the pet's movement patterns.

[1145] The activity detection sensor records the pet's activity level and uses an accelerometer to measure how active the pet is.

[1146] 2. Information processing device:

[1147] Data transmission: The terminal sends the collected data to the server (information processing device) at regular intervals (e.g., every 5 minutes). This process is carried out using a secure protocol (e.g., TLS).

[1148] Data Analysis and Anomaly Detection: The server analyzes the received data and detects anomalies. First, it converts the data into a parseable format (e.g., CSV or JSON). Next, it cleans the data and extracts features (such as exercise levels, meal frequency, and toilet usage frequency). Based on these features, it applies machine learning algorithms to detect abnormal behavior.

[1149] 3. Notice:

[1150] Anomaly Notification: The server notifies the terminal of any detected anomalies. The terminal displays this information to the user through a user interface (such as a smartphone app). For example, an alert might be displayed stating, "Your pet is behaving unusually. Please check on it."

[1151] 4. Gathering feedback:

[1152] User Behavior and Feedback: Users check the status of their pets based on the notified abnormal information and take appropriate measures as needed. Users provide feedback on their observations and actions taken through the application.

[1153] Model retraining: The terminal sends the provided feedback information to the server, which then retrains the anomaly detection model based on this feedback. This improves the system's accuracy and enables more precise anomaly detection in the future.

[1154] Specific example

[1155] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws, and this video data is sent from the device to a server. The server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. As a result, the device sends a notification to the user stating, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian." Based on this information, the user can check their pet's paws and, if there is a problem, take it to a veterinarian, enabling early detection and treatment.

[1156] An example of a prompt statement is: "Use a system that collects pet behavior data using cameras and sensors and analyzes it on a server to monitor the pet's health in real time and detect abnormalities."

[1157] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1158] Step 1:

[1159] Data collection

[1160] The device collects behavioral data in real time using location sensors (e.g., GPS sensors) and activity sensors (e.g., accelerometers) attached to the pet. Inputs include signals from the sensors (location information, acceleration data). These sensor signals are converted into digital data and stored in a temporary buffer memory within the device. Outputs consist of a set of location information and activity data.

[1161] Specific actions:

[1162] The GPS sensor collects location information every second.

[1163] The accelerometer records the pet's movements every 0.1 seconds.

[1164] The collected data is temporarily stored in a buffer.

[1165] Step 2:

[1166] Data transmission

[1167] The terminal sends data stored in a buffer to the server at regular intervals (e.g., every 5 minutes). The input is location and momentum data stored in the buffer, and the output is data packets sent to the server. Secure protocols such as TLS are used for transmission, and the order of the data is maintained by timestamping each data packet.

[1168] Specific actions:

[1169] The data stored in the buffer is read and sent to the server using the TLS protocol.

[1170] After transmission is complete, the buffer is cleared to prepare for the next data collection.

[1171] Step 3:

[1172] Data Analysis

[1173] The server receives and analyzes data sent from the terminal. The input is location and activity data sent from the terminal, and the output is analyzed features (e.g., activity level, number of meals, frequency of toilet use, etc.) and anomaly detection results. First, the raw data is converted into an analyzable format (e.g., CSV or JSON), and outliers and missing values ​​are checked and cleaned. Then, specific features are extracted, and abnormal behavior is analyzed using a machine learning algorithm.

[1174] Specific actions:

[1175] Convert the raw data to a different format and clean the data as needed.

[1176] Extract features and organize them into an analyzable format.

[1177] Machine learning algorithms are applied to detect anomalies.

[1178] Step 4:

[1179] Abnormal notification

[1180] The server notifies the terminal of any anomalies detected. The input consists of the analyzed data and the results of the anomaly detection, while the output is the notification message sent to the terminal. The notification includes the type, date, time, and frequency of the detected anomaly.

[1181] Specific actions:

[1182] A notification message is constructed based on the detected anomaly data.

[1183] Send a message to the device and display it on the user interface.

[1184] Step 5:

[1185] Feedback Collection

[1186] The device collects user feedback and sends it to the server. The input is feedback information provided by the user, and the output is feedback data sent to the server. The feedback includes detailed information such as the user's observations of the pet's condition, the actions taken, and the results.

[1187] Specific actions:

[1188] Collect feedback through the user interface.

[1189] The collected feedback is sent to the server.

[1190] Step 6:

[1191] Model Retraining

[1192] The server retrains its machine learning algorithm based on feedback received from the user. The input consists of feedback data and historical behavioral data, and the output is an updated anomaly detection model. This improves the accuracy of anomaly detection.

[1193] Specific actions:

[1194] Integrate feedback data with past behavioral data.

[1195] Retrain the anomaly detection model and apply the new model.

[1196] (Application Example 1)

[1197] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1198] Properly monitoring a pet's health while it is traveling and detecting abnormalities early is crucial for ensuring their safety and well-being. Conventional pet health monitoring systems are typically used within the home or in specific locations, and monitoring during travel has not been considered. As a result, it has been difficult to ascertain a pet's health status in a timely manner, especially during long journeys. Furthermore, while a quick and accurate response is required when an abnormality is detected in a pet, there is a challenge in that appropriate means of response are not readily available while the pet is traveling.

[1199] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1200] In this invention, the server includes means for collecting pet behavior data using sensors and cameras, means for temporarily storing the collected behavior data, means for transmitting the stored behavior data to the server, means for analyzing the transmitted behavior data and applying a machine learning model to detect anomalies, means for notifying the terminal of the detected anomaly, means for displaying the notified anomaly information and providing the user with appropriate countermeasures, means for collecting user feedback and transmitting it to the server, means for retraining the machine learning model using the transmitted feedback, means for collecting pet behavior data using sensors and cameras mounted in the autonomous vehicle, means for processing the data in real time and detecting anomalies in cooperation with the system in the autonomous vehicle, and means for notifying the vehicle's dashboard display and smartphone when an anomaly is detected. This makes it possible to appropriately monitor the health status of pets even while they are in transit and to detect anomalies early.

[1201] "Pet behavioral data" refers to information related to a pet's movement patterns, activity level, and health status.

[1202] A "sensor" is a device that captures environmental information or physical quantities and converts them into electrical signals, etc. Specific examples include GPS sensors and accelerometers.

[1203] A "camera" is a device that captures images and videos, and includes 360-degree cameras and night vision cameras.

[1204] "Means of temporary storage" refers to memory or buffers used to retain data for a certain period of time.

[1205] "Means of sending to the server" refers to a communication module that has the function of transferring collected data to a computer in a remote location.

[1206] A "machine learning model" refers to an algorithm that uses past data to recognize patterns and then analyzes and predicts new data.

[1207] "Means for detecting anomalies" refers to a function that uses machine learning models to identify anomalies that deviate from normal behavioral patterns.

[1208] "Means of notifying the terminal" refers to communication devices or applications that have the function of transmitting anomaly detection information to the user.

[1209] A "dashboard display" refers to a display device installed inside an autonomous vehicle, used to visually display notification information and warnings.

[1210] A "smartphone" refers to a mobile device equipped with multiple functions, such as voice communication, data communication, and application execution.

[1211] "User feedback" refers to responses and information provided by users regarding the system's operation and notifications.

[1212] "Methods for processing data in real time" refer to functions that perform analysis and decision-making almost simultaneously with the moment data is collected.

[1213] An "autonomous vehicle" refers to a vehicle that operates autonomously without the intervention of a driver.

[1214] This invention relates to a system designed for the early detection of pet health issues and illnesses. This system uses multiple sensors and cameras to collect behavioral data for monitoring pet health within autonomous vehicles, and analyzes this data on a server to detect abnormalities. The specific configuration and operation of this system are described in detail below.

[1215] Data collection

[1216] The terminal (a platform installed inside the autonomous vehicle) collects behavioral data in real time via sensors and cameras attached to the pet. For example, a GPS sensor tracks the pet's movement patterns, and an accelerometer records the pet's activity level. Cameras installed inside the vehicle capture the pet's actions as video, and the pet's movements are captured as video data. This data is temporarily stored in a buffer within the terminal. Specifically, if the pet is frequently moving or active inside the vehicle, its movements are recorded with a 360-degree camera.

[1217] Data transmission

[1218] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (TLS) to ensure data consistency and security. The data is time-stamped to accurately preserve its order.

[1219] Data analysis and anomaly detection

[1220] The server analyzes the received data. First, it converts the data into a format that can be analyzed, and then cleans it by checking for outliers and missing values. Next, it extracts features to compare with past behavioral data. Specific features include exercise levels, meal frequency, and toilet usage frequency. Based on these features, a generative AI model is used to detect anomalies. When an abnormal behavioral pattern is detected, information that has been judged as abnormal is generated.

[1221] Anomaly notification and countermeasures provided

[1222] The server notifies the terminal of any detected anomalies. The terminal provides an interface to notify the user (pet owner) of this information, displaying it via voice, dashboard display, smartphone, etc. For example, if an anomaly is detected, an alert such as "Your pet is behaving unusually. Please check on them" will be displayed on the vehicle's dashboard display and smartphone.

[1223] User behavior and feedback

[1224] Based on the notified abnormal information, users can check their pet's condition and take appropriate action. For example, if the camera captures the pet frequently licking its front paws, the user can use that information to check the pet's paws and take it to the veterinarian if any abnormalities are found. Users can also provide feedback on their observations and the actions taken through the application.

[1225] Gathering feedback and retraining the model

[1226] The device sends feedback information provided by the user to the server. The server then retrains the anomaly detection model based on this feedback information, improving the system's accuracy. This makes it possible to monitor the pet's health more accurately.

[1227] As a concrete example, by using a prompt statement such as, "Write a program that analyzes sensor and camera data in real time to monitor the health of pets and detect abnormalities," the system's operation can be made to work exactly as specified.

[1228] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1229] Step 1:

[1230] The device collects behavioral data in real time using sensors and cameras attached to the pet. The input is raw data from the sensors and camera, and the output is data temporarily stored in a buffer within the device. Specifically, it simultaneously acquires the pet's location information from the GPS sensor, activity level from the accelerometer, and video data from the camera, and records each of these in the buffer.

[1231] Step 2:

[1232] The terminal sends data stored in a buffer to the server at regular intervals. The input here is the behavioral data stored in the buffer, and the output is the data sent to the server. The TLS protocol is used for this transmission, ensuring data consistency and security. Specifically, once data has accumulated in the buffer for a certain period, it is uploaded to the server in a single batch.

[1233] Step 3:

[1234] The server analyzes the received data. The input is behavioral data sent from the terminal, and the output is data converted into a parseable format. Specifically, after parsing the data, it checks for outliers and missing values ​​and performs cleaning. It also analyzes timestamp information to maintain the integrity of time-series data.

[1235] Step 4:

[1236] The server extracts features from the analyzed data and uses them to apply a generative AI model to detect anomalies. The input in this process is cleaned data, and the output is the anomaly detection result. Specifically, it extracts features such as exercise volume, meal frequency, and toilet usage frequency, and inputs these into the generative AI model to detect anomaly patterns.

[1237] Step 5:

[1238] The server notifies the terminal of any detected anomalies. The input is the result of the anomaly detection, and the output is the data for notification. Specifically, as soon as an anomaly is detected, the server sends that information to the terminal in real time and prepares to issue a warning to the user.

[1239] Step 6:

[1240] The terminal notifies the user of the received anomaly information. The input here is the anomaly information sent from the server, and the output is the notification to the user. Specifically, it displays the anomaly details and corrective actions using the vehicle's dashboard display or the smartphone's notification function.

[1241] Step 7:

[1242] The user checks the pet's condition based on the notified abnormal information and takes appropriate action as needed. The input is notifications from the terminal, and the output is user feedback. Specifically, the user actually observes the pet's condition and reports the results to the system via the terminal.

[1243] Step 8:

[1244] The terminal sends user-provided feedback information to the server. The input is the user's feedback, and the output is the data sent to the server. Specifically, it collects feedback information and sends it to the server along with a timestamp.

[1245] Step 9:

[1246] The server retrains the machine learning model using the feedback it receives. The input is the feedback data sent by the user, and the output is the new machine learning model with improved accuracy. Specifically, it incorporates the collected feedback data as part of the training data and updates the generative AI model.

[1247] By following the steps outlined above, it is possible to effectively implement a system that monitors a pet's health in real time and detects abnormalities early.

[1248] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1249] This invention is a system for supporting pet health management, which collects and analyzes pet behavioral data in real time and detects abnormalities. Furthermore, the invention incorporates an emotion engine that recognizes the user's emotions, and adds a function to adjust notifications and countermeasures based on the user's emotions. The specific configuration and embodiments of this system are described below.

[1250] Data collection

[1251] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. The GPS sensor tracks the pet's movement patterns, and the accelerometer records the pet's activity level. Simultaneously, a camera is used to capture the pet's behavior as video and detect specific actions (e.g., walking, sitting, licking). This data is temporarily stored in a buffer within the device.

[1252] Data transmission and analysis

[1253] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security.

[1254] The server converts the received data into an analyzable format, checks for outliers and missing values, and performs data cleaning. Next, it extracts features for comparison with past pet behavior data and uses a machine learning model to detect anomalies.

[1255] Using an Emotion Engine

[1256] This system includes an emotion engine that recognizes the user's emotions. The terminal captures the user's voice and facial expressions and sends that data to the server.

[1257] The server analyzes this data and applies emotion recognition algorithms. For example, it identifies whether the user is stressed or relaxed. This analysis result is taken into consideration when notifying users of abnormal information about their pet.

[1258] Anomaly notification and countermeasures provided

[1259] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the pet's abnormal behavior. The content and method of notification are adjusted based on the analysis results of the emotion engine. For example, if the user is stressed, the notification will be expressed in a softer tone, while if they are relaxed, more detailed information will be provided.

[1260] The device displays this information to the user through a user interface and provides appropriate countermeasures. For example, a message such as, "Your pet is behaving unusually. Please check on it," might be displayed.

[1261] User behavior and feedback

[1262] Users check notifications, observe their pet's condition, and take necessary actions. For example, they might check their pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken.

[1263] Gathering feedback and retraining the model

[1264] The device sends user-provided feedback information and sentiment data to the server.

[1265] Based on this feedback, the server retrains its anomaly detection model and emotion recognition algorithm to improve the system's accuracy and personalization. This allows for more accurate monitoring of pet health and the provision of optimal notifications and actions tailored to the user's emotional state.

[1266] As a concrete example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. Simultaneously, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and don't panic." The user can then use this information to check on their pet's condition and, if necessary, take it to a veterinarian, enabling early detection and treatment.

[1267] The following describes the processing flow.

[1268] Step 1:

[1269] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. Specifically, the GPS sensor periodically records the pet's location coordinates, and the accelerometer monitors the pet's activity level (e.g., intensity and frequency of movement). Furthermore, a camera is used to capture the pet's behavior as video, and specific actions (e.g., walking, sitting, licking) are automatically detected.

[1270] Step 2:

[1271] The device temporarily stores the collected behavioral data in a buffer. The buffer is time-stamped to accurately preserve the order of the data. For example, daily movement data, activity level data, and video data are stored in chronological order.

[1272] Step 3:

[1273] The terminal sends data stored in a buffer to the server at regular intervals (e.g., every minute). Data communication is securely performed using the TLS (Transport Layer Security) protocol, ensuring data consistency and security.

[1274] Step 4:

[1275] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and normalizes the sensor's numerical data. For example, the video data is converted into JPEG frames, and the acceleration data is scaled within a specified range.

[1276] Step 5:

[1277] The server performs data cleaning. It checks for outliers and missing values ​​and removes or fills in invalid data. For example, it identifies and processes abnormal spikes in activity level data or missing consecutive frames.

[1278] Step 6:

[1279] The server extracts features to compare with past pet behavior data stored in the database. For example, it calculates features such as daily exercise, number of meals, and frequency of litter box use.

[1280] Step 7:

[1281] The server uses a machine learning model to analyze features and detect abnormal pet behavior. For example, it can detect abnormal behavior such as "constantly licking its paws" by comparing it to normal behavior patterns. The machine learning model is trained based on past training data.

[1282] Step 8:

[1283] The device captures the user's voice and facial expressions and sends that data to a server. For example, it uses the smartphone's microphone and camera to record audio and video while the user is observing their pet.

[1284] Step 9:

[1285] The server analyzes the transmitted audio and video data and uses an emotion engine to recognize the user's emotions. For example, it can determine whether the user is stressed or relaxed based on changes in voice tone and facial expressions.

[1286] Step 10:

[1287] The server determines the content and method of notification based on the detected anomaly information and the results of the emotion engine's analysis. For example, if the user is feeling stressed, it will generate a gentle message such as, "Your pet is licking its paws repeatedly. We understand you are worried, but please take your time and check on it carefully."

[1288] Step 11:

[1289] The device displays the generated notification content to the user and suggests appropriate actions. For example, it might display a message via the smartphone's notification function saying, "Your pet is behaving unusually. We recommend consulting a veterinarian."

[1290] Step 12:

[1291] The user checks the notification, observes the pet's condition, and takes appropriate action (e.g., takes it to the vet). They then input feedback about the observation results and the actions taken into the application.

[1292] Step 13:

[1293] The device sends user-provided feedback and sentiment data to the server. For example, it sends text information entered by the user into the application and diagnostic results.

[1294] Step 14:

[1295] The server uses feedback information and sentiment data to retrain machine learning models and sentiment recognition algorithms, improving the system's accuracy. This allows the system to more accurately monitor pet health and provide optimal notifications and actions tailored to the user's emotional state.

[1296] (Example 2)

[1297] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1298] A current challenge in pet health management is the difficulty in quickly and accurately detecting abnormal behavior in pets and notifying users of this information in an appropriate manner. Furthermore, because the user's emotional state is not taken into consideration, the content of the notifications may cause excessive stress to the user. To solve these problems, this invention aims to provide a system that collects and analyzes not only pet behavior data but also user emotional data, and provides optimal notifications and countermeasures tailored to individual situations.

[1299] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1300] In this invention, the server includes means for analyzing pet behavior data and applying a machine learning model to detect anomalies, means for analyzing user emotion data and applying an emotion recognition algorithm, and means for adjusting anomaly notifications based on the user's emotional state. This enables high-precision detection of pet behavioral anomalies and the provision of appropriate notifications and countermeasures that take into account the user's emotional state.

[1301] "Pet behavioral data" refers to data that shows a pet's movement patterns, activity levels, and specific behaviors (e.g., walking, sitting, licking).

[1302] A "sensor" is a device used to measure the state of the environment or objects. Specifically, this includes GPS sensors and accelerometers.

[1303] A "camera" is a device used to capture video data.

[1304] A "buffer" is a memory area used to temporarily store data.

[1305] A "server" is a computer system that processes data over a network.

[1306] A "machine learning model" is a model built using algorithms to make predictions and classifications based on data.

[1307] "Emotional data" refers to data that indicates the emotional state of a user, obtained from their voice and facial expressions.

[1308] An "emotion identification algorithm" is an algorithm that analyzes emotional data to identify a user's emotional state (e.g., stress, relaxation).

[1309] An "abnormality notification" is a notification sent to the user when an abnormality is detected in the pet's behavior.

[1310] "User feedback" refers to information about observations and actions taken by users that they provide to the system.

[1311] "Retraining" is the process of improving a machine learning model using newly collected data.

[1312] This invention is a system for supporting pet health management, specifically by collecting and analyzing pet behavioral data in real time and detecting abnormalities. Furthermore, the invention incorporates an emotion engine that recognizes the user's emotions, and adds a function to adjust notifications and countermeasures based on the user's emotions.

[1313] Data collection

[1314] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. The GPS sensor tracks the pet's movement patterns, and the accelerometer records the pet's activity level. Simultaneously, a camera is used to capture the pet's behavior as video and detect specific actions (e.g., walking, sitting, licking). This data is temporarily stored in a buffer within the device.

[1315] Data transmission and analysis

[1316] The terminal sends data stored in a buffer to the server at regular intervals. The transmission process uses a secure protocol (e.g., TLS) to ensure data consistency and security.

[1317] The server converts the received data into an analyzable format, checks for outliers and missing values, and performs data cleaning. Next, it extracts features for comparison with past pet behavior data and uses a machine learning model to detect anomalies.

[1318] Using an Emotion Engine

[1319] This system includes an emotion engine that recognizes the user's emotions. The terminal captures the user's voice and facial expressions and sends that data to the server.

[1320] The server analyzes this data and applies emotion recognition algorithms. For example, it identifies whether the user is stressed or relaxed. This analysis is taken into consideration when notifying users of abnormal information about their pet.

[1321] Anomaly notification and countermeasures provided

[1322] The server notifies the terminal of any detected anomalies. This notification includes detailed information such as the type, date, time, and frequency of the pet's abnormal behavior. The content and method of notification are adjusted based on the analysis results of the emotion engine. For example, if the user is stressed, the notification will be expressed in a softer tone, while if they are relaxed, more detailed information will be provided.

[1323] The device displays this information to the user through a user interface and provides appropriate countermeasures. For example, a message such as, "Your pet is behaving unusually. Please check on it," might be displayed.

[1324] User behavior and feedback

[1325] Users check notifications, observe their pet's condition, and take necessary actions. For example, they might check their pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken.

[1326] Gathering feedback and retraining the model

[1327] The device sends user-provided feedback information and sentiment data to the server.

[1328] Based on this feedback, the server retrains its anomaly detection model and emotion recognition algorithm to improve the system's accuracy and personalization. This allows for more accurate monitoring of pet health and the provision of optimal notifications and actions tailored to the user's emotional state.

[1329] Specific example

[1330] For example, consider a scenario where a camera captures a pet frequently licking its front paws. This video data is sent from the device to a server, where the server analyzes the "continuous paw licking" behavior pattern and determines it to be abnormal. Simultaneously, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and don't panic." The user can then use this information to check on their pet's condition and, if necessary, take it to a veterinarian, enabling early detection and treatment.

[1331] Example of a prompt:

[1332] "Please explain how the system works to detect abnormal pet behavior and provide appropriate notifications and countermeasures based on the user's emotions."

[1333] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1334] Step 1:

[1335] Data collection

[1336] The device collects behavioral data in real time using GPS and accelerometer sensors attached to the pet. First, it obtains the pet's current location from the GPS sensor every second, and measures the acceleration along the X, Y, and Z axes in milliseconds using the accelerometer. Furthermore, it captures the pet's behavior as video at 30 frames per second using a camera, and uses an image recognition algorithm to identify specific actions. This data is temporarily stored in a buffer. Sensor data and video data are collected as input and stored in the buffer as output.

[1337] Step 2:

[1338] Data transmission

[1339] The device sends activity data stored in a buffer to the server every 30 seconds. The transmitted data includes GPS data, acceleration data, and video data. The transmission process uses the TLS protocol to ensure data security and consistency. It takes data from the buffer as input and sends it to the server as output.

[1340] Step 3:

[1341] Data Analysis

[1342] The server converts the received data into a format that can be analyzed. Specifically, it divides the video data into frames and identifies the pet's movements using an image recognition algorithm. GPS data and acceleration data are synchronized using timestamps, and outliers and missing values ​​are filtered out. Features are extracted using the cleaned data. The server takes the received data as input and obtains the analyzed data and features as output.

[1343] Step 4:

[1344] Collection and transmission of emotional data

[1345] The device uses the front camera and microphone to collect user emotion data. The camera captures the user's facial expressions, and the microphone records their voice. This data is temporarily stored in a buffer and sent to the server at regular intervals. User facial expression data and voice data are taken as input, stored in a buffer as output, and sent to the server.

[1346] Step 5:

[1347] Emotion analysis

[1348] The server analyzes the received emotional data. It uses a facial recognition algorithm to analyze the user's facial features and identify emotions. It also uses a speech recognition algorithm to analyze audio data and estimate emotions from voice tone and speed. Using these analysis results, it determines whether the user is stressed or relaxed. It takes the received emotional data as input and obtains the emotional state as output.

[1349] Step 6:

[1350] Anomaly detection

[1351] The server uses a machine learning model to detect anomalies based on features. For example, if a pet repeatedly performs the same action within a specific range (e.g., continuously licking its front paws), this is detected as an anomaly. The server classifies the type of anomaly and associates it with specific behavioral patterns. It takes cleaned data and features as input and obtains anomaly detection results as output.

[1352] Step 7:

[1353] Abnormal notification

[1354] The server generates a notification message based on the anomaly detection results. This message includes the type of abnormal behavior, date and time, frequency, and content based on the sentiment analysis results. It provides gentle information if the user is stressed, and detailed information if they are relaxed. The terminal displays the received notification to the user. For example, it displays the notification content using smartphone push notifications or the interface of a dedicated app. It takes anomaly detection results and sentiment analysis results as input, and displays the generated notification to the user as output.

[1355] Step 8:

[1356] Collecting user feedback

[1357] The user checks the notification and actually observes the pet's condition. For example, they check the pet's paws and consult a veterinarian if they find any abnormalities. Afterwards, they provide feedback through the application regarding their observations and the actions taken. The application takes the notification content as input and provides information regarding observations and actions taken as output.

[1358] Step 9:

[1359] Feedback analysis and model retraining

[1360] The device sends user-provided feedback and sentiment data to the server. The server uses this feedback to retrain its anomaly detection model and sentiment recognition algorithm. Specifically, it uses the feedback information as labeled data to improve the accuracy and personalization of the machine learning model. It retrieves the feedback information sent as input and obtains the improved anomaly detection model and sentiment recognition algorithm as output.

[1361] (Application Example 2)

[1362] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1363] Conventional pet health management systems have the functionality to collect and analyze pet behavior data in real time, but they do not provide notifications or countermeasures that take into account the user's emotional state. As a result, when the user is stressed, the notification content may be inappropriate, and there may be cases where an abnormality in the pet cannot be addressed quickly. The present invention aims to provide a more effective pet health management system that takes the user's emotions into consideration.

[1364] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting and transmitting user emotion data to the server, means for analyzing the transmitted emotion data and applying an emotion identification algorithm, and means for adjusting countermeasures based on the detected anomaly information and the user's emotional state and notifying the terminal. This makes it possible to monitor the pet's health status in real time while providing appropriate notifications and countermeasures according to the user's emotional state.

[1365] "Behavioral data" refers to data that shows a pet's movement patterns and activity levels, and is collected using GPS sensors and accelerometers.

[1366] A "sensor" is a device that detects physical information about the environment or objects, such as a GPS sensor or an accelerometer, and outputs that information as data.

[1367] A "camera" is a device used to capture a pet's behavior as video, and it is equipped with the function to detect specific actions.

[1368] A "server" refers to a computer system used to analyze collected data, detect anomalies, and analyze user emotions.

[1369] A "machine learning model" includes data analysis algorithms used to detect anomalies based on pet behavior data.

[1370] "Emotion recognition" refers to an algorithm that identifies a user's emotional state by analyzing their voice and facial expression data.

[1371] An "abnormal" behavior refers to a state in which a pet exhibits behavior that is clearly different from its normal behavioral patterns, suggesting a potential health problem.

[1372] "Notifications" refer to information sent to inform users about detected anomalies or measures taken based on the user's emotional state.

[1373] "Countermeasures" include instructions that show how to respond to detected anomalies or what actions to take.

[1374] "User feedback" refers to information provided by users regarding their pet's condition and the measures taken, which is used to improve the system's performance.

[1375] "Retraining" refers to the process of improving the performance of machine learning models and sentiment recognition algorithms using newly collected data.

[1376] This invention is a system for supporting pet health management, which uses sensors and cameras to collect and analyze pet behavior data in real time. Furthermore, by collecting and analyzing user emotional data, the system can provide notifications and countermeasures based on the user's emotional state. Based on this embodiment, a method for effectively managing a pet's health will be described.

[1377] System Configuration

[1378] The system mainly consists of the following components:

[1379] Sensors: These are GPS sensors and accelerometers attached to the pet, which are used to obtain the pet's location information and movement data.

[1380] Camera: A high-resolution camera that captures specific actions of your pet (e.g., walking, sitting, licking).

[1381] Devices (smartphones, tablets, etc.): Temporarily store collected data and send it to the server as needed.

[1382] Server: A computer system that performs data analysis, analyzing collected data using machine learning models and sentiment recognition algorithms.

[1383] User interface: An application used to notify users of abnormal information and countermeasures.

[1384] System operation

[1385] 1. Data Collection: Pet behavior data collected by sensors (e.g., GPS data, accelerometer data) and behavioral video data captured by cameras are transmitted to the device. The device also collects user voice and facial expression data.

[1386] 2. Data transmission: The collected data is sent to the server at regular intervals using a secure protocol.

[1387] 3. Analysis and Anomaly Detection: The server converts the received data into an analyzable format and uses machine learning models to detect abnormal pet behavior. Furthermore, it analyzes the user's emotional state using an emotion recognition algorithm.

[1388] 4. Provision of Notifications and Countermeasures: If an anomaly is detected, the server sends the most appropriate notification and countermeasure to the terminal based on the anomaly information and the user's emotional state. The terminal then notifies the user of this through the user interface.

[1389] 5. User Response and Feedback: Users will review notifications and take appropriate action regarding their pet's abnormal behavior. They will provide feedback on their observations and the measures taken through the application.

[1390] 6. System Improvement: The server retrains machine learning models and sentiment recognition algorithms based on user feedback to improve the system's accuracy and personalization.

[1391] Specific example

[1392] For example, if the camera captures a pet frequently licking its front paws, the video data is sent from the device to the server. The server analyzes the data and determines that the "continuous paw licking" behavior is abnormal. At the same time, if the emotion engine detects that the user is experiencing stress from their voice and facial expressions, the abnormality notification will include a gentle message such as, "Your pet is continuously licking its paws. There may be something wrong with its paws. We recommend consulting a veterinarian. However, for now, please remain calm and do not panic." This allows the user to take appropriate action and effectively manage their pet's health.

[1393] Example of a prompt

[1394] "What should I do if my pet is frequently exhibiting signs of illness?"

[1395] This system allows for real-time monitoring of pet health and prompt, appropriate action. Furthermore, by providing notifications and solutions tailored to the user's emotional state, it reduces the user's mental burden, enabling more confident pet health management.

[1396] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1397] Step 1:

[1398] The device receives behavioral data collected by GPS and accelerometer sensors attached to the pet. It also collects video data of the pet's behavior captured by a camera. Simultaneously, the device collects the user's voice and facial expression data. In this step, the inputs are data from sensors and cameras, and the user's voice and facial expression data, and the output is raw data that is temporarily stored.

[1399] Step 2:

[1400] The terminal sends the collected data to the server at regular intervals. Communication is performed using a secure protocol (e.g., TLS). In this step, there is temporarily stored data as input, and data is generated to be sent to the server as output.

[1401] Step 3:

[1402] The server converts the received data into a parseable format. It performs data cleaning, checking for outliers and missing values, and formatting the data. In this step, the raw data received is the input, and the cleaned data is produced as the output.

[1403] Step 4:

[1404] The server inputs the formatted data into a machine learning model to analyze the pet's behavior patterns. An anomaly detection model is applied to detect abnormal behavior. In this step, the input is cleaned data, and the output is the anomaly detection result.

[1405] Step 5:

[1406] The server analyzes the user's voice and facial expression data and applies an emotion recognition algorithm to determine whether the user is stressed or relaxed. In this step, the user's voice and facial expression data are the input, and the emotion recognition result is generated as the output.

[1407] Step 6:

[1408] The server integrates anomaly detection results and sentiment identification results to generate appropriate notification content and countermeasures. It adjusts the text and wording of the notification based on the user's emotional state. In this step, the input is anomaly detection results and sentiment identification results, and the output is the adjusted notification content.

[1409] Step 7:

[1410] The terminal receives notifications from the server and displays them through the user interface. The user reviews the notification content and takes appropriate action. In this step, the input is the adjusted notification content, and the output is the notification sent to the user.

[1411] Step 8:

[1412] The user checks the pet's condition and takes necessary actions based on the notification. Feedback on observations and actions taken is provided through the application. In this step, the user's observation and action information is the input, and feedback information is generated as the output.

[1413] Step 9:

[1414] The server receives feedback from the user and uses it to retrain the machine learning model and sentiment recognition algorithm. In this step, the received feedback information is taken as input, and the updated model is produced as output.

[1415] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1416] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1417] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1418] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1419] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1420] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1421] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1422] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1423] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1424] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1425] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1426] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1427] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1428] 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.

[1429] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1430] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1431] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1432] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1433] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1434] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1435] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1436] The following is further disclosed regarding the embodiments described above.

[1437] (Claim 1)

[1438] A means of collecting pet behavior data using sensors and cameras,

[1439] A means of temporarily storing the collected behavioral data,

[1440] A means of sending stored behavioral data to a server,

[1441] A means of analyzing transmitted behavioral data and applying a machine learning model to detect anomalies,

[1442] A means of notifying the terminal of information about detected anomalies,

[1443] A means of displaying the notified abnormal information and providing the user with appropriate countermeasures,

[1444] A means of collecting user feedback and sending it to a server,

[1445] A means of retraining a machine learning model using the submitted feedback,

[1446] A system that includes this.

[1447] (Claim 2)

[1448] The system according to claim 1, characterized in that the sensor and camera include a GPS sensor and an accelerometer.

[1449] (Claim 3)

[1450] The system according to claim 1, characterized in that the machine learning model is trained on past behavioral data and health data of the pet.

[1451] "Example 1"

[1452] (Claim 1)

[1453] A means of collecting pet behavior data using a detection device,

[1454] A means of temporarily storing the collected behavioral data,

[1455] Means for transmitting stored behavioral data to an information processing device,

[1456] A means for analyzing transmitted behavioral data and applying machine learning algorithms to detect anomalies,

[1457] A means of notifying a display terminal of detected anomaly information,

[1458] A means of displaying the notified abnormal information and providing users with appropriate countermeasures,

[1459] A means for collecting user feedback and transmitting it to an information processing device,

[1460] A means of retraining a machine learning algorithm using the submitted feedback,

[1461] A system that includes this.

[1462] (Claim 2)

[1463] The system according to claim 1, characterized in that the detection device includes a position detection sensor and a momentum detection sensor.

[1464] (Claim 3)

[1465] The system according to claim 1, characterized in that the machine learning algorithm is trained on past behavioral data and health data of the pet.

[1466] "Application Example 1"

[1467] (Claim 1)

[1468] A means of collecting pet behavior data using sensors and cameras,

[1469] A means of temporarily storing the collected behavioral data,

[1470] A means of sending stored behavioral data to a server,

[1471] A means of analyzing transmitted behavioral data and applying a machine learning model to detect anomalies,

[1472] A means of notifying the terminal of information about detected anomalies,

[1473] A means of displaying the notified abnormal information and providing the user with appropriate countermeasures,

[1474] A means of collecting user feedback and sending it to a server,

[1475] A means of retraining a machine learning model using the submitted feedback,

[1476] A means of collecting pet behavior data using sensors and cameras installed in an autonomous vehicle,

[1477] A means of processing data in real time and detecting anomalies in conjunction with the system inside the autonomous vehicle,

[1478] A means of notifying the vehicle's dashboard display and smartphone when an abnormality is detected,

[1479] A system that includes this.

[1480] (Claim 2)

[1481] The system according to claim 1, characterized in that the sensor and camera include a GPS sensor and an accelerometer.

[1482] (Claim 3)

[1483] The system according to claim 1, characterized in that the machine learning model is trained on past behavioral data and health data of the pet.

[1484] "Example 2 of combining an emotion engine"

[1485] (Claim 1)

[1486] A means of collecting pet behavior data using sensors and cameras,

[1487] A means of temporarily storing the collected behavioral data,

[1488] A means of sending stored behavioral data to a server,

[1489] A means of analyzing transmitted behavioral data and applying a machine learning model to detect anomalies,

[1490] A means of collecting user sentiment data and sending it to a server,

[1491] A means for analyzing transmitted emotion data and applying an emotion recognition algorithm,

[1492] A means of adjusting abnormal notifications based on the user's emotional state,

[1493] A means of notifying the terminal of information about detected anomalies,

[1494] A means of displaying the notified abnormal information and providing the user with appropriate countermeasures,

[1495] A means of collecting user feedback and sending it to a server,

[1496] A means of retraining a machine learning model using the submitted feedback,

[1497] A system that includes this.

[1498] (Claim 2)

[1499] The system according to claim 1, characterized in that the sensor and camera include a GPS sensor and an accelerometer.

[1500] (Claim 3)

[1501] The system according to claim 1, characterized in that the machine learning model is trained on past behavioral data and health data of the pet.

[1502] "Application example 2 when combining with an emotional engine"

[1503] (Claim 1)

[1504] A means of collecting pet behavior data using sensors and cameras,

[1505] A means of temporarily storing the collected behavioral data,

[1506] A means of sending stored behavioral data to a server,

[1507] A means of analyzing transmitted behavioral data and applying a machine learning model to detect anomalies,

[1508] A means of collecting user sentiment data and sending it to a server,

[1509] A means for analyzing transmitted emotion data and applying an emotion recognition algorithm,

[1510] A means of adjusting countermeasures based on detected anomaly information and the user's emotional state, and notifying the device,

[1511] A means of displaying the notified abnormal information and providing the user with appropriate countermeasures,

[1512] A means of collecting user feedback and sending it to a server,

[1513] A means of retraining machine learning models and sentiment recognition algorithms using the submitted feedback,

[1514] A system that includes this.

[1515] (Claim 2)

[1516] The system according to claim 1, characterized in that the sensor and camera include a GPS sensor and an accelerometer.

[1517] (Claim 3)

[1518] The system according to claim 1, characterized in that the machine learning model is trained on past behavioral data and health data of the pet. [Explanation of Symbols]

[1519] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting pet behavior data using sensors and cameras, A means of temporarily storing the collected behavioral data, A means of sending stored behavioral data to a server, A means of analyzing transmitted behavioral data and applying a machine learning model to detect anomalies, A means of notifying the terminal of information about detected anomalies, A means of displaying the notified abnormal information and providing the user with appropriate countermeasures, A means of collecting user feedback and sending it to a server, A means of retraining a machine learning model using the submitted feedback, A system that includes this.

2. The system according to claim 1, characterized in that the sensor and camera include a GPS sensor and an accelerometer.

3. The system according to claim 1, characterized in that the machine learning model is trained based on past behavioral data and health data of the pet.

Images