Childcare support system

Abstract

This system provides a childcare support system that is suitable for understanding information based on the individual characteristics of the children being cared for. [Solution] The generating AI server 100 registers reference information, including the child's body temperature, visual examination results, and parental comments, in the knowledge base 12. Then, it inputs a request to the AI ​​model 10 that includes a request to generate information related to the child's health management, and in response to the request, it retrieves information output from the AI ​​model 10 by referring to the reference information in the knowledge base 12. As a result, information related to the child's health management based on the child's body temperature, visual examination results, and parental comments can be obtained, making it possible to grasp information based on the child's individuality and needs, compared to conventional methods.

Description

Technical Field

[0001] The present invention relates to a system for supporting childcare, and more particularly to a childcare support system suitable for grasping information based on the individuality of the child in care, etc.

Background Art

[0002] The declining birthrate problem is in a serious situation. For this reason, the government is promoting various measures to address the declining birthrate and aiming to recover the birthrate. In 2023, the "Agency for Children and Families" was established, and in the future, positive measures to address the declining birthrate such as increasing family benefits and improving childcare are planned. Also, the "Universal Nursery System" implemented since 2024 is a nursery system that allows parents to leave their pre-school children at a nursery even if they are not working. Currently, basically, nurseries can only be used by families that have received recognition of the need for childcare, such as when the guardian is working or there are special circumstances such as illness or caregiving. In contrast, the "Universal Nursery System" was formulated so that "anyone" can use a nursery on an hourly basis regardless of the employment of the guardian. In the model project that has already been implemented since last year, facilities such as nurseries with available spaces and certified children's gardens are the target facilities. While the need from guardians is very high for this system, many opinions from nurseries view the implementation with concern, such as not being able to keep up with the system for securing childcare workers.

[0003] From such a background, DX (Digital Transformation) for nurseries has been developed and put into the market. As DX for nurseries, for example, the technology described in Non-Patent Document 1 is known.

[0004] The technology described in Non-Patent Document 1 has functions such as management of arrival and departure from the nursery, communication with guardians, creation of lesson plans and logs, shift management, billing management, childcare documentation, contact books, calculation of extended childcare fees, temperature measurement and infectious disease checks, near misses, growth and health records, emergency communication, questionnaires, photo sharing and sales, management of multiple facilities, security, meal and menu management, bus operation management, etc.

Prior Art Documents

[0005] [Non-Patent Document 1] "Childcare ICT System CoDMON | For Nurseries and Educational Facilities | No. 1 Market Share in the Industry", [online], August 9, 2024, CoDMON Co., Ltd., [Searched August 9, 2024], Internet<URL:https: / / www.codmon.com / > [Overview of the project] [Problems that the invention aims to solve]

[0006] Under the system where any child can attend daycare, childcare workers are responsible for different children each day, making it difficult to provide childcare that is tailored to each child's individuality and needs. For example, when creating a daily childcare plan, it is desirable to take into account the individuality of each child on that day, but because different children attend each day, it is difficult to take into account the individuality of each child on that day, making it impossible to create an appropriate childcare plan. This problem cannot be solved even with the technology described in Non-Patent Document 1.

[0007] Therefore, the present invention has been made in view of the unresolved problems of the conventional technology, and aims to provide a childcare support system suitable for grasping information based on the individuality of the person being cared for. [Means for solving the problem]

[0008] [Invention 1] To achieve the above objective, the childcare support system of Invention 1 includes: registration means for registering activity information relating to the number of times a childcare worker has contact with a childcare recipient, or activity information relating to the amount of activity of the childcare recipient or the number of times they have contact with other people, in a knowledge base that can be referenced by an AI model; input means for inputting a request to the AI ​​model that includes a request to generate information relating to the childcare recipient; and acquisition means for obtaining information output from the AI ​​model in response to the request by having the AI ​​model refer to the activity information in the knowledge base.

[0009] In this configuration, activity information is registered in the knowledge base via the registration means. Then, a request including a request to generate information about the person being cared for is input to the AI ​​model via the input means, and the information output from the AI ​​model in response to the request is retrieved via the acquisition means. At this time, the AI ​​model refers to the activity information in the knowledge base.

[0010] Here, the input means include, for example, directly inputting a request into the AI ​​model, or indirectly inputting it into the AI ​​model through processing, functions, devices, networks, or other means. The same applies hereafter to the childcare support system of Invention 2.

[0011] Furthermore, means of acquisition include, for example, directly acquiring output information of the AI ​​model, or indirectly acquiring output information of the AI ​​model through processing, functions, devices, networks, or other means. The same applies hereafter to the childcare support system of Invention 2.

[0012] Furthermore, activity information relating to activity levels includes, for example, activity information relating to one or more activity levels, activity information relating to the activity level itself, or activity information relating to activity level statistics. Similarly, activity information relating to the number of contacts includes, for example, activity information relating to one or more contact counts, activity information relating to the number of contacts itself, or activity information relating to the number of contacts statistics. The same applies hereafter to the childcare support system of Invention 2.

[0013] Furthermore, requests for generating information about the person being cared for include, for example, requests for generating answer information to questions about the person being cared for. The same applies hereafter to the childcare support system of Invention 2.

[0014] Furthermore, generation requests include, for example, those that explicitly request the generation of information about the person being cared for, and those that indirectly request the generation of information about the person being cared for. An explicit request would be, for example, "Please generate information about the person being cared for." An indirect request would be, for example, to request the generation of answer information to a question about the person being cared for, by making a question about the person being cared for the generation request. This is because the AI ​​model will generate answer information if a question about the person being cared for is input as a prompt. The same applies hereafter to the childcare support system of Invention 2.

[0015] Furthermore, activity information and requests can be composed of, for example, vector data or any other format. The same applies hereafter to the childcare support system of Invention 2.

[0016] Furthermore, the knowledge base stores activity information by any means and at any time. It may store activity information in advance, or it may store activity information through external input or other means during the operation of the system without storing it in advance.

[0017] Furthermore, this system may be implemented as a single device, apparatus, terminal, or other device, or as a network system in which multiple devices, apparatus, terminals, or other devices are connected in a communicative manner. In the latter case, each component may belong to any of the multiple devices, as long as they are connected in a communicative manner. The same applies hereafter to the childcare support system of Invention 2.

[0018] [Invention 2] Furthermore, the childcare support system of Invention 2 includes an input means for inputting a request to the AI ​​model that includes a request to generate information about the child, which includes activity information relating to the number of times a caregiver has contact with a child, or activity information relating to the amount of activity of the child, or the number of times the child has contact with other people; and an acquisition means for acquiring information output from the AI ​​model in response to the request.

[0019] With such a configuration, a request including activity information and a generation request for information regarding the ward is input into the AI model by the input means, and information output from the AI model for the request is acquired by the acquisition means.

[0020] 〔Invention 3〕Furthermore, in the childcare support system of Invention 3, in the childcare support system according to any one of Inventions 1 and 2, the request for generating information regarding the ward is a request for generating information regarding the plan, evaluation, or improvement of group childcare implemented collectively for a plurality of the wards.

[0021] With such a configuration, a request including a generation request for information regarding the plan, evaluation, or improvement of group childcare is input into the AI model by the input means, and information output from the AI model for the request is acquired by the acquisition means.

[0022] Here, examples of the plan, evaluation, or improvement of group childcare include the plan, evaluation, or improvement of the wards who received group childcare, or the plan, evaluation, or improvement of the childcare workers who conducted group childcare.

[0023] 〔Invention 4〕Furthermore, in the childcare support system of Invention 4, in the childcare support system according to any one of Inventions 1 and 2, the request for generating information regarding the ward is a request for generating information regarding the plan, evaluation, or improvement of the measures implemented for the development, growth, or advancement of the ward.

[0024] With such a configuration, a request including a generation request for information regarding the plan, evaluation, or improvement of the measures is input into the AI model by the input means, and information output from the AI model for the request is acquired by the acquisition means.

[0025] Here, examples of the plan, evaluation, or improvement of the measures include the plan, evaluation, or improvement of the childcare worker for the measures implemented by the childcare worker, or the plan, evaluation, or improvement of the ward who utilized the measures.

[0026] [[Invention 5]] Further, in the childcare support system of Invention 5, in the childcare support system of either Invention 1 or 2, the request for generating information regarding the child in care is a request for generating safety management information regarding the safety management of the child in care.

[0027] With such a configuration, a request including a request for generating safety management information is input into the AI model by the input means, and information output from the AI model in response to the request is acquired by the acquisition means. [[Advantages of the Invention]]

[0028] As described above, according to the childcare support system of Invention 1 or 2, information based on the number of contacts between the childcare worker and the child in care, or the amount of activity of the child in care or the number of contacts with others can be obtained. Therefore, compared with the prior art, information based on the personality and needs of the child in care can be grasped.

[0029] Furthermore, according to the childcare support system of Invention 3, information regarding the planning, evaluation or improvement of group childcare can be obtained, so that the planning, evaluation or improvement of group childcare becomes easy.

[0030] Furthermore, according to the childcare support system of Invention 4, information regarding the planning, evaluation or improvement of measures can be obtained, so that the planning, evaluation or improvement of measures becomes easy.

[0031] Furthermore, according to the childcare support system of Invention 5, safety management information regarding the safety management of the child in care can be obtained, so that safety management becomes easy. [[Brief Description of the Drawings]]

[0032] [Figure 1] It is a diagram showing an overview of the service according to the present embodiment. [Figure 2] It is a block diagram showing the configuration of the network system according to the present embodiment. [Figure 3] It is a diagram showing the hardware configuration of the generation AI server 100. [Figure 4] This is a flowchart showing the knowledge data registration process. [Figure 5] This diagram shows the data structure of the play menu information. [Figure 6] This is a flowchart showing the process for generating response information. [Figure 7] This diagram shows the data structure used when generating planning information for integrated childcare. [Figure 8] This diagram shows the data structure used when generating planning information for integrated childcare. [Figure 9] This diagram shows the data structure used when generating evaluation information for integrated childcare. [Figure 10] This diagram shows the data structure used when generating evaluation information for integrated childcare. [Figure 11] This diagram shows the data structure used when generating information on improving centralized childcare services. [Figure 12] This diagram shows the data structure used when generating planning information for developmental measures. [Figure 13] This diagram shows the data structure used when generating evaluation information for developmental policies. [Figure 14] This diagram shows the data structure used when generating child safety management information. [Figure 15] This diagram shows the data structure used when generating health management information for children. [Figure 16] This diagram shows the data structure used when generating information on children's meal management. [Figure 17] This diagram shows the data structure used when generating emergency response information related to children's health. [Modes for carrying out the invention]

[0033] The embodiments of the present invention will be described below. Figures 1 to 17 show these embodiments.

[0034] First, the configuration of this embodiment will be described. [Service Overview] Figure 1 is a diagram showing an overview of the service according to this embodiment.

[0035] As shown in Figure 1, the generation AI server 100 is configured to include an AI (Artificial Intelligence) model 10 and a knowledge base 12 that registers data that the AI ​​model 10 references for inference.

[0036] The generating AI server 100 receives basic information, environmental information, qualitative information, quantitative information, or image information from a childcare terminal 200 installed in a childcare center, a childcare worker terminal 210 used by childcare workers, or a wearable device 220 worn by a child attending the childcare center (hereinafter simply referred to as "child"), and registers this received information in the knowledge base 12. Knowledge data such as childcare guidelines are also registered in the knowledge base 12.

[0037] When the generation AI server 100 receives a request from a childcare worker terminal 210 or the like that includes a request to generate information about a child, it causes the AI ​​model 10 to refer to the data in the knowledge base 12 in response to the received request, and causes the AI ​​model 10 to generate response information related to the generation request. The generated response information is then sent to the childcare worker terminal 210 or the like.

[0038] On the childcare worker terminal 210, etc., childcare workers can ask questions via UI (User Interface) or voice. The childcare worker terminal 210, etc. sends a request to the generation AI server 100 that includes a request to generate answer information for the question entered via UI or voice. Upon receiving the answer information in response to the request, the AI ​​server displays the received answer information.

[0039] The generating AI server 100, through predetermined processing, causes the AI ​​model 10 to refer to data from the knowledge base 12 and generate information about children in the AI ​​model 10. Based on the generated information, it then sends breaking news, signals, alerts, and other notifications to childcare worker terminals 210, etc.

[0040] When the childcare worker terminal 210 or similar device receives a breaking news report, signal, alert, or other notification, it displays the received notification.

[0041] [Network System] Next, the configuration of the network system according to this embodiment will be described.

[0042] Figure 2 is a block diagram showing the configuration of the network system according to this embodiment. As shown in Figure 2, the Internet 199 is connected to a generation AI server 100, one or more nursery terminals 200, one or more childcare worker terminals 210, and one or more wearable devices 220. Regarding allocation, for example, the nursery terminals 200 are allocated per nursery, the childcare worker terminals 210 are allocated per childcare worker, and the wearable devices 220 are allocated per childcare worker and per child.

[0043] [Generating AI Server 100] Next, we will explain the configuration of the generation AI server 100. Figure 3 shows the hardware configuration of the generation AI server 100.

[0044] As shown in Figure 3, the generation AI server 100 consists of a CPU (Central Processing Unit) 30 that controls calculations and the entire system based on a control program, a ROM (Read Only Memory) 32 that stores the control program for the CPU 30 in a predetermined area, a RAM (Random Access Memory) 34 for storing data read from the ROM 32 and other memory, as well as calculation results necessary for the calculation process of the CPU 30, and an I / F (Interface) 38 that mediates data input and output to external devices. These components are connected to each other and enable data exchange via a bus 39, which is a signal line for data transfer.

[0045] I / F38 is connected to an external device, which includes an input device 40 consisting of a keyboard and mouse that can input data as a human interface, a storage device 42 that stores data and tables as files, a display device 44 that displays a screen based on an image signal, and a signal line for connecting to the Internet 199.

[0046] The memory device 42 stores the AI ​​model 10 and the knowledge base 12. AI Model 10 is an AI model trained on a large dataset and is a highly versatile model capable of performing various tasks. For example, AI Model 10 can employ a Large Language Model. A Large Language Model is a deep learning model that pre-trains a language model, which models human speech based on its probability of occurrence, on a vast amount of data. When a prompt is input, the Large Language Model statistically infers the probability of generating the next word from the sentence contained in the input prompt and outputs the inference result. For example, publicly known techniques described on the internet sites "https: / / chatgpt-lab.com / n / n418d3aa56f0b" and "https: / / agirobots.com / chatgpt-mechanism-and-problem / " can be used as a Large Language Model. More specifically, for example, Titan Text G1 - Express, Titan Text G1 - Lite, Titan Image Generator G1, Titan Embeddings G1 - Text, Titan Embeddings Text V2, Titan Multimodal Embeddings G1, Claude, Claude Instant, Claude 3 Sonnet, Claude 3 Haiku, Claude 3 Opus, Jurassic-2 Mid, Jurassic-2 Ultra, Command, Command Light, Command R, Command R+, Embed English, Embed Multilingual, Llama 2 Chat 13B, Llama 2 Chat 70B, Llama 2 13B, Llama 2 70B, Llama 3 8b Instruct, Llama 3 70b Instruct, Mistral 7B Instruct, Mixtral 8X7B Instruct, Mistral Large, and Stable Diffusion XL can be adopted.

[0047] Knowledge base 12 can register information received from the nursery terminal 200, the childcare worker terminal 210, or the wearable device 220. Knowledge base 12 also registers knowledge data such as childcare guidelines, including (1) information on childcare guidelines for nursery schools, (2) information on laws and other rules or standards related to childcare, (3) information on childcare manuals, or (4) emergency response information. Emergency response information includes, for example, (1) emergency response information related to health (e.g., sudden illness, allergy information, injury, difficulty breathing), (2) emergency response information related to the environment (e.g., fire, earthquake or natural disaster, power outage, gas leak), (3) emergency response information related to security (e.g., intruder intrusion, kidnapping or missing person), or (4) other emergency response information (e.g., outbreak of infection, sudden illness or shortage of staff).

[0048] [Nursery school terminal 200 and nursery teacher terminal 210] Next, the configuration of the nursery terminal 200 and the childcare worker terminal 210 will be explained.

[0049] The nursery terminal 200 and the childcare worker terminal 210 are configured with the same hardware configuration as the generation AI server 100. The nursery terminal 200 and the childcare worker terminal 210 can be configured as smartphones or tablet devices.

[0050] [Wearable device 220] Next, we will describe the configuration of the wearable device 220.

[0051] The wearable device 220 is a device that can be worn on the wrist, arm, neck, or other parts of a childcare worker or child, and is configured to have the same functions as a general computer with a CPU, ROM, RAM, and I / F connected via a bus.

[0052] The wearable device 220 is equipped with a vital sensor and an accelerometer, and calculates the wearer's activity level (e.g., steps taken, distance traveled, calories burned, heart rate) based on the detection results of the vital sensor and accelerometer, and transmits the activity information, including the calculated activity level, to the generating AI server 100.

[0053] The wearable device 220 is equipped with an RFID (Radio Frequency Identification) tag and reader. The RFID reader reads the RFID tags of other wearable devices 220 that are within a predetermined range (for example, within the range of close proximity or individual distance). Based on the detection results of the RFID reader, the wearer calculates the number of times they have come into contact with other wearers (hereinafter simply referred to as "number of contacts") and transmits activity information, including the calculated number of contacts, to the generating AI server 100. The number of contacts with others serves as an indicator of the child's social interaction (how much they interacted with other children and caregivers).

[0054] The wearable device 220 is equipped with a radio wave strength sensor and, based on the detection results of the radio wave strength sensor, transmits activity information, including the radio wave strength used for BLE (Bluetooth Low Energy) communication with a specific wearable device 220 (hereinafter referred to as "BLE strength"), to the generating AI server 100. For example, if a wearable device 220 worn by a childcare worker is set as a specific wearable device 220, activity information including the BLE strength for the wearable device 220 worn by the childcare worker can be transmitted to the generating AI server 100. This activity information allows the system to determine how far away the child is from the childcare worker.

[0055] Next, the operation of this embodiment will be described. [Knowledge Data Registration] First, let's explain the case where the generating AI server 100 registers knowledge data.

[0056] Figure 4 is a flowchart showing the knowledge data registration process. The CPU 30 consists of an MPU (Micro-Processing Unit), etc., and starts a predetermined program stored in a predetermined area of ​​the ROM 32, and executes the knowledge data registration process shown in the flowchart of Figure 4 according to that program. The knowledge data registration process is executed when a request including a registration request for registering knowledge data (hereinafter simply referred to as "registration request") is received, and when it is executed in the CPU 30, it first proceeds to step S100, as shown in Figure 4.

[0057] In step S100, activity information and other information to be registered are received, the process moves to step S102, where the received activity information is stored in the storage device 42, and the process moves to step S104.

[0058] In step S104, the received activity information is converted into vector data that the AI ​​model 10 can refer to. Vector data can be generated by embedding, a technique that converts data including text, images, and audio into numerical vectors.

[0059] Next, the process moves to step S106, where the converted vector data is registered in the knowledge base 12. At this time, the vector data is registered in association with identification information in order to allow the AI ​​model 10 to refer to specific data in the knowledge base 12. The request includes a request to refer to the knowledge base 12 and the identification information, allowing the AI ​​model 10 to refer to the data in the knowledge base 12 that corresponds to the identification information.

[0060] Identification information includes, for example, (1) identification information for identifying a nursery school or a group thereof, (2) identification information for identifying a nursery teacher or a group thereof, (3) identification information for identifying a child or a group thereof, (4) identification information for identifying basic information, environmental information, qualitative information, quantitative information, image information or other information, (5) identification information for identifying childcare provided to multiple children at once (hereinafter referred to as "group childcare"), (6) identification information for identifying the date, time, duration, location or other information related to the implementation of group childcare, (7) identification information for identifying a child or a group thereof who received group childcare, or (8) identification information for identifying measures implemented for the development, growth or development of one or more children (for example, creation of a childcare curriculum, changes to the outdoor or indoor layout of the nursery school) (hereinafter referred to as "developmental measures"). This allows the AI ​​model 10 to reference data from the knowledge base 12 by specifying, for example, (1) data corresponding to a nursery school or a group thereof, (2) data corresponding to a nursery teacher or a group thereof, (3) data corresponding to a child or a group thereof, (4) data corresponding to basic information, environmental information, qualitative information, quantitative information, image information or other information, (5) data corresponding to group childcare, (6) data corresponding to the date, time, location or other information related to the implementation of group childcare, (7) data corresponding to a child or a group thereof who received group childcare, or (8) data corresponding to developmental measures, using identification information.

[0061] Once the process in step S106 is completed, the series of processes ends. [Activity Information Registration] In the wearable device 220, at predetermined intervals (e.g., every 10 minutes) or at predetermined or arbitrary timings, a request including activity information regarding the wearer's activity level, number of contacts, and BLE intensity, as well as a registration request, is sent to the generating AI server 100, and the activity information is registered in the knowledge base 12. The activity information is registered in association with the identification information of the childcare worker or child related to that activity information, the date and time information on when the activity information was acquired, or the location information on when the activity information was acquired. As a result, activity information for multiple children is registered as knowledge data.

[0062] [Registration of basic information, etc.] On the childcare center terminal 200 or the childcare worker terminal 210 (hereinafter referred to as "childcare worker terminal 210, etc."), a request including basic information, environmental information, qualitative information, quantitative information, image information, or other information and a registration request is sent to the generation AI server 100 at predetermined intervals (for example, every 10 minutes) or at predetermined or arbitrary timings, and the basic information, etc. is registered in the knowledge base 12. The basic information, etc. is registered in association with the identification information of the childcare center, childcare worker, or child related to the basic information, etc., the date and time information on when the basic information, etc. was acquired, or the location information on when the basic information, etc. was acquired.

[0063] Basic information may include, for example, (1) information about the nursery school's name, address, number of childcare workers, number of children, and other attributes; (2) information about the childcare workers' names, ages, genders, affiliated nursery schools, and other attributes; (3) information about the children's names, ages, genders, height, weight, body temperature, allergy information, vaccination history, medical history, affiliated nursery schools, and other attributes; (4) information about comments from childcare workers or guardians regarding the child; or (5) information about emergencies involving the child.

[0064] Environmental information includes, for example, (1) information on temperature, humidity, atmospheric pressure, wind direction, wind speed, light intensity, noise, intensity of sunlight (e.g., ultraviolet rays), intensity of electromagnetic waves, concentration of carbon dioxide, weather, or smell in the nursery or other childcare environment, or (2) information on developmental measures. These are obtained by installing temperature sensors, etc., in the nursery or other childcare environment, and the detection results of the temperature sensors, etc., are acquired by the childcare worker terminal 210, etc., and a request including the acquired detection results as environmental information is sent to the generation AI server 100.

[0065] Qualitative information (information that is difficult to quantify) includes, for example, (1) information about behavior (e.g., how to play, how to interact with friends, behavioral characteristics), (2) information about emotions (e.g., expression of joy, anger, sadness, etc., expression of achievement, expression of failure, emotional stability), (3) information about thinking (e.g., problem-solving methods, curiosity, imagination), (4) information about language (e.g., vocabulary, speaking style, comprehension), and (5) information about social skills (e.g., cooperativeness, empathy, awareness of rules). It should be noted that information that partially overlaps with basic information may be considered qualitative information, and information that does not overlap with basic information may also be considered qualitative information.

[0066] Quantitative information (information that can be quantified) includes, for example, (1) information related to development, growth, or development (e.g., height, weight, vocabulary size, motor skills such as walking, running, and jumping), (2) information related to lifestyle habits (e.g., diet, amount of food eaten, sleep duration, playtime), (3) information related to learning (e.g., number of correct and incorrect answers to questions, number of books that can be read), and (4) information related to health (e.g., body temperature, number of visits to medical institutions, medication use, allergy information). In addition, quantitative information may include information that partially overlaps with activity information, basic information, and environmental information, or information that does not overlap with activity information, basic information, and environmental information.

[0067] Figure 5 shows the data structure of the play menu information. The quantitative information also includes play menu information related to children's play and their activity levels. The play menu information is structured as a table with one record registered for each type of play, as shown in Figure 5, for example. Each record includes fields to register the name of the play, a summary of the play, activity elements, variations of the play, and activity levels (for example, an estimated value for 20 minutes of activity (e.g., number of steps)). In the example in Figure 5, seven menus are registered as children's park play activities.

[0068] Image information includes, for example, corresponding to the above examples of qualitative and quantitative information, image information relating to behavior, emotions, thoughts, language, social skills, development, growth, lifestyle, learning, and health.

[0069] [Answer information generation] Next, we will explain the case where the generation AI server 100 generates response information using the AI ​​model 10.

[0070] Figure 6 is a flowchart showing the response information generation process. The CPU 30 starts a predetermined program stored in a predetermined area of ​​the ROM 32 and executes the response information generation process shown in the flowchart of Figure 6 according to that program. The response information generation process is executed when a request to generate information about a child is received, and when it is executed by the CPU 30, it first proceeds to step S200, as shown in Figure 6.

[0071] In step S200, a request is received to generate answer information. The request includes a request to generate answer information and parameters. A generation request may include, for example, a request to generate answer information for a question about a child, or a request to generate other information about a child. Parameters may include, for example, a reference request to refer to all or specific data in Knowledge Base 12, or, in the case of a reference request to refer to specific data, identification information to identify that data.

[0072] Next, the process moves to step S202, where a prompt is generated to instruct the AI ​​model 10 based on the generation request included in the received request. If the generation request is in the form of a prompt, the generation request is retrieved from the received request, and the retrieved generation request is used as the prompt.

[0073] Next, the process moves to step S204, where the generated prompt is input to the AI ​​model 10. Here, if the received request includes a reference request, the reference request is input to the AI ​​model 10; if it includes both a reference request and identification information, both the reference request and identification information are input to the AI ​​model 10.

[0074] When a prompt is entered, AI model 10 infers the answer information from the entered prompt. As a result of this inference, AI model 10 provides answer information that matches the training data and generation request of AI model 10.

[0075] When a prompt and a reference request are input, the AI ​​model 10 references all the data in the knowledge base 12 based on the input reference request and infers answer information from the input prompt. As a result of the AI ​​model 10's inference, answer information that fits the AI ​​model 10's training data, all the data in the knowledge base 12, and the generation request is obtained.

[0076] When AI model 10 receives prompts, reference requests, and identification information, it refers to specific data in knowledge base 12 based on the entered reference requests and identification information, and infers answer information from the entered prompts. As a result of this inference, AI model 10 provides answer information that matches the AI ​​model 10's training data, specific data in knowledge base 12, and the generation request.

[0077] Next, the process moves to step S206 to obtain the inference results of the AI ​​model 10, then to step S208 to send the response information, including the obtained inference results, to the requesting terminal, and the series of processes ends.

[0078] [1. Generation of planning information for integrated childcare] Figures 7 and 8 show the data structure when generating a plan for integrated childcare. Figure (a) shows the data structure of the request, Figure (b) shows the data structure of the information in the knowledge base 12 that the AI ​​model 10 references, and Figure (c) shows the data structure of the response information from the generating AI server 100.

[0079] The first example shows how to generate a plan for afternoon-time integrated childcare from morning-time information. When a childcare worker wants to obtain planning information regarding a group childcare plan, which is based on the number of times the childcare worker interacts with the children and the amount and number of times the children are active, they input a request to generate planning information on the childcare worker terminal 210, etc. When a request to generate planning information is input on the childcare worker terminal 210, etc., a request including the input generation request and parameters is sent to the generation AI server 100. The request consists of, for example, the generation request "Morning data... Please consider measures," and the parameters, a reference request and identification information "Childcare center: Midori no Oka Nursery School... Information regarding the childcare guidelines of the nursery school," as shown in Figure 7(a). Of these parameters, "Reference Information 1" is information from the knowledge base 12 regarding the time the children are at the nursery in the morning, and indicates that the AI ​​model 10 should refer to (1) the child's identification information, (2) the child's name, (3) activity information 1 showing the measured values ​​when the children had free play in the park for one hour in the morning, (4) activity information 2 showing the measured values ​​when the children had story time in the nursery for one hour in the morning, and (5) visual inspection comments. Activity Information 1 contains information about the child's activity level (e.g., number of steps), the number of times the childcare worker interacts with the child, and the number of times the child interacts with the child. Activity Information 2 is similar. Therefore, the information in the knowledge base 12 that the AI ​​model 10 references is structured as shown in Figure 7(b), for example. Furthermore, "Reference Information 2" indicates that the AI ​​model 10 will reference information from the knowledge base 12 regarding the childcare guidelines for daycare centers.

[0080] Then, in response to this request, planning information is obtained from the generating AI server 100. The planning information is configured as shown in Figure 7(c).

[0081] The second example shows how to generate a plan for the remaining 40 minutes of integrated childcare after 20 minutes of free play in the park.

[0082] When a request for generating plan information is input to the childcare worker terminal 210, a request including the input generation request and parameters is sent to the generation AI server 100. The request consists of, for example, the generation request "Regarding 1 hour of play in the park... Please also create the reason," and the parameters, a reference request and identification information "Nursery school: Midori no Oka Nursery School... Play menu information," as shown in Figure 8(a). Of these parameters, "Reference Information 1" is information from the knowledge base 12 regarding free play in the park as an outing for a walk, and indicates that the AI ​​model 10 should refer to (1) the child's identification information, (2) the child's name, (3) Activity Information 1 showing the measured values ​​when free play in the park was performed for 20 minutes from the start, (4) Activity Information 2 showing the actual values ​​when free play in the park was performed for 20 minutes from the start in the past, and (5) BLE intensity. Activity Information 1 is information regarding the child's activity level (e.g., number of steps) and the number of times the child made contact. Activity Information 2 is similar. Therefore, the information in the knowledge base 12 that the AI ​​model 10 references is configured as shown in Figure 8(b), for example. Furthermore, "Reference Information 2" indicates that the AI ​​model 10 is being instructed to reference the game menu information from the knowledge base 12.

[0083] Then, in response to this request, planning information is obtained from the generating AI server 100. The planning information is configured as shown in Figure 8(c).

[0084] [2. Generation of evaluation information for integrated childcare] Figures 9 and 10 show the data structure when generating evaluation information for integrated childcare. Figure (a) shows the data structure of the request, Figure (b) shows the data structure of the information in the knowledge base 12 that the AI ​​model 10 references, and Figure (c) shows the data structure of the response information from the generating AI server 100.

[0085] The first example shows how to generate evaluation information for a group childcare session when "tag" is played after 20 minutes of free play in a park.

[0086] When a childcare worker wants to obtain evaluation information regarding the evaluation of group childcare, based on the number of times the childcare worker interacts with the child and the amount of activity the child is engaged in, they input a request to generate evaluation information on the childcare worker terminal 210 or the like. When the childcare worker terminal 210 or the like receives a request to generate evaluation information, a request including the input generation request and parameters is sent to the generation AI server 100. The request consists of, for example, the generation request "Event start 20 minutes... Please evaluate," and the parameters, a reference request and identification information "Childcare center: Midori no Oka Nursery School... Information regarding the childcare center's childcare guidelines," as shown in Figure 9(a). Of these parameters, "Reference Information 1" is information from the knowledge base 12 regarding free play in the park as part of an outing for a walk, and indicates that the AI ​​model 10 should refer to (1) the child's identification information, (2) the child's name, (3) activity information 1 showing the measured values ​​when free play in the park was conducted for 20 minutes from the start, (4) activity information 2 showing the measured values ​​when free play in the park was conducted for 60 minutes from the start, and (5) childcare worker comments. Activity Information 1 contains information about the child's activity level (e.g., number of steps) and the number of times the childcare worker interacts with the child. Activity Information 2 is similar. Therefore, the information in the knowledge base 12 that the AI ​​model 10 references is structured as shown in Figure 9(b), for example. Furthermore, "Reference Information 2" indicates that the AI ​​model 10 will reference information from the knowledge base 12 regarding the childcare guidelines for daycare centers.

[0087] In response to this request, evaluation information is obtained from the generating AI server 100. The evaluation information is structured as shown in Figure 9(c).

[0088] The second example involves generating evaluation information regarding the evaluation of group childcare provided to a specific child over a certain period of time.

[0089] When a childcare worker wants to obtain evaluation information regarding the evaluation of group childcare, based on the number of times the childcare worker interacts with the children and the amount and number of times the children are active, they input a request to generate evaluation information on the childcare worker terminal 210, etc. When a request to generate evaluation information is input on the childcare worker terminal 210, etc., a request including the input generation request and parameters is sent to the generation AI server 100. The request consists of, for example, the generation request "Please evaluate the childcare guidelines for the nursery school," and the parameters, a reference request and identification information "Nursery school: Midori no Oka Nursery School... Information regarding the childcare guidelines for the nursery school." Of these parameters, "Reference Information 1" is information from the knowledge base 12, specifically information about free play in the park and story time inside the nursery school as part of outings, and indicates that the AI ​​model 10 should refer to (1) the child's identification information, (2) the child's name, (3) the child's age (in months), (4) activity information 1 showing the measured values ​​when free play in the park was performed for one hour, and (5) activity information 2 showing the measured values ​​when story time was performed for one hour inside the nursery school. Activity Information 1 contains information about the child's activity level (e.g., number of steps), the number of times the caregiver interacted with the child, the number of times the child interacted with the child, and comments from the caregiver. Activity Information 2 is similar. Therefore, the information in the knowledge base 12 that the AI ​​model 10 references is structured as shown in Figure 10(b), for example. Furthermore, "Reference Information 2" indicates that the AI ​​model 10 will reference information from the knowledge base 12 regarding the childcare guidelines for daycare centers.

[0090] In response to this request, evaluation information is obtained from the generating AI server 100. The evaluation information is structured as shown in Figure 10(c).

[0091] [3. Generating information to improve integrated childcare] Figure 11 shows the data structure when generating information to improve integrated childcare. Figure (a) shows the data structure of the request, Figure (b) shows the data structure of the information in the knowledge base 12 that the AI ​​model 10 refers to, and Figure (c) shows the data structure of the response information from the generating AI server 100.

[0092] When a childcare worker wants to obtain improvement information regarding improvements to group childcare, based on the number of times the childcare worker interacts with the child and the amount of activity the child is engaged in, they input a request to generate improvement information at the childcare worker terminal 210 or the like. When a request to generate improvement information is input at the childcare worker terminal 210 or the like, a request including the input generation request and parameters is sent to the generation AI server 100. The request consists of, for example, the generation request "Please provide the measurement values ​​showing the results of execution Do, and... response," and the parameters, a reference request and identification information "Childcare center: Midori no Oka Nursery School... play menu information." Of these parameters, "reference information 1" is information from the knowledge base 12 regarding free play in the park as part of an outing for a walk, and indicates that the AI ​​model 10 should refer to (1) the child's identification information, (2) the child's name, (3) activity information 1 showing the measurement values ​​when free play in the park was conducted for 20 minutes from the start, (4) activity information 2 showing the measurement values ​​when free play in the park was conducted for 60 minutes from the start, and (5) childcare worker comments. Activity Information 1 contains information about the child's activity level (e.g., number of steps) and the number of times the childcare worker interacts with the child. Activity Information 2 is similar. Therefore, the information in the knowledge base 12 that the AI ​​model 10 references is structured as shown in Figure 11(b), for example. Furthermore, "Reference Information 2" indicates that the AI ​​model 10 will reference information from the knowledge base 12 regarding the childcare guidelines for daycare centers.

[0093] Then, in response to this request, improvement information is obtained from the generating AI server 100. The improvement information is structured as shown in Figure 11(c).

[0094] [4. Generation of planning information for developmental measures] Figure 12 shows the data structure when generating development policy planning information. Figure (a) shows the data structure of the request, Figure (b) shows the data structure of the information in the knowledge base 12 that is referenced by the AI ​​model 10, and Figure (c) shows the data structure of the response information from the generating AI server 100.

[0095] When a childcare worker wants to obtain planning information regarding developmental measures, such as a plan based on the number of times the childcare worker interacts with the child and the amount and number of times the child is active, they input a request to generate planning information at the childcare worker terminal 210, etc. When a request to generate planning information is input at the childcare worker terminal 210, etc., a request including the input generation request and parameters is sent to the generation AI server 100. The request consists of, for example, the generation request "Please propose the introduction of an atelier," and the parameters, a reference request and identification information "Nursery school: Midori no Oka Nursery School... Information regarding the nursery school's childcare guidelines," as shown in Figure 12(a). Of these parameters, "Reference Information 1" is information from the knowledge base 12 regarding activities within the nursery school, and indicates that the AI ​​model 10 should refer to (1) the child's identification information, (2) the child's name, and (3) activity information showing the measured values ​​when activities were conducted within the nursery school before the introduction of the atelier. The activity information is information regarding the child's amount of activity (e.g., number of steps), the number of times the childcare worker interacts with the child, and the number of times the child interacts with the child. Therefore, the information in the knowledge base 12 that the AI ​​model 10 references is structured as shown in Figure 12(b), for example. Furthermore, "Reference Information 2" indicates that the AI ​​model 10 is being instructed to reference information from the knowledge base 12 that pertains to childcare guidelines for daycare centers.

[0096] Then, in response to this request, planning information is obtained from the generating AI server 100. The planning information is structured as shown in Figure 12(c).

[0097] [5. Generation of evaluation information for developmental policies] Figure 13 shows the data structure when generating evaluation information for developmental measures. Figure (a) shows the data structure of the request, Figure (b) shows the data structure of the information in the knowledge base 12 that is referenced by the AI ​​model 10, and Figure (c) shows the data structure of the response information from the generating AI server 100.

[0098] When a childcare worker wants to obtain evaluation information related to the evaluation of developmental measures, based on the number of times the childcare worker interacts with the child and the amount and number of times the child is active, they input a request to generate evaluation information on the childcare worker terminal 210, etc. When a request to generate evaluation information is input on the childcare worker terminal 210, etc., a request including the input generation request and parameters is sent to the generation AI server 100. The request consists of, for example, the generation request "Please evaluate the atelier, ..." and the parameters, a reference request and identification information "Nursery school: Midori no Oka Nursery School... Information on the nursery school's childcare guidelines". Of these parameters, "Reference Information 1" is information from the knowledge base 12 regarding activities within the nursery school, and indicates that the AI ​​model 10 should refer to (1) the child's identification information, (2) the child's name, (3) activity information 1 showing the measured values ​​when activities were conducted within the nursery school before the introduction of the atelier, (4) activity information 2 showing the measured values ​​when activities were conducted within the nursery school after the introduction of the atelier, and (5) childcare worker comments. Activity Information 1 contains information about the child's activity level (e.g., number of steps), the number of times the childcare worker interacts with the child, and the number of times the child interacts with the child. Activity Information 2 is similar. Therefore, the information in the knowledge base 12 that the AI ​​model 10 references is structured as shown in Figure 13(b), for example. Furthermore, "Reference Information 2" indicates that the AI ​​model 10 will reference information from the knowledge base 12 regarding the childcare guidelines for daycare centers.

[0099] In response to this request, evaluation information is obtained from the generating AI server 100. The evaluation information is structured as shown in Figure 13(c).

[0100] [6. Generating information for improving developmental policies] When a childcare worker wants to obtain improvement information regarding developmental measures, based on the number of times the childcare worker interacts with the children and the amount and number of times the children interact with them, they input a request for the generation of improvement information using the childcare worker terminal 210 or the like. Here, the request, the information from the knowledge base 12 to be referenced, and the response information from the generation AI server 100 can be configured as a data structure similar to that shown in Figure 11.

[0101] [7. Generation of safety management information] Figure 14 shows the data structure when generating child safety management information. Figure (a) shows the data structure of the request, Figure (b) shows the data structure of the information in the knowledge base 12 that is referenced by the AI ​​model 10, and Figure (c) shows the data structure of the response information from the generating AI server 100.

[0102] When a childcare worker wants to obtain safety management information for one or more children based on the number of times the childcare worker has contact with them and the amount and number of times the children have been active, they input a request to generate safety management information on the childcare worker terminal 210, etc. When a request to generate safety management information is input on the childcare worker terminal 210, etc., a request including the input generation request and parameters is sent to the generation AI server 100. The request consists of, for example, the generation request "A risk has occurred to the child... Please compare and comment," and the parameters, a reference request and identification information "Nursery school: Midori no Oka Nursery School... BLE strength," as shown in Figure 14(a). The "reference information" among these parameters is information from the knowledge base 12 regarding free play in the park as part of an outing, and indicates that the AI ​​model 10 should refer to (1) the child's identification information, (2) the child's name, (3) activity information 1 showing the measured values ​​when free play was performed in the park, (4) activity information 2 showing the actual values ​​when free play was performed in the park in the past, and (5) BLE strength. Activity Information 1 contains information about the child's activity level (e.g., number of steps), the number of times the caregiver interacts with the child, and the number of times the child interacts with the child. Activity Information 2 is similar. Therefore, the information in the knowledge base 12 that the AI ​​model 10 references is structured as shown in Figure 14(b), for example.

[0103] In response to this request, safety management information is obtained from the generating AI server 100. The safety management information is structured as shown in Figure 14(c).

[0104] [8. Generation of health management information] Figure 15 shows the data structure when generating child health management information. Figure (a) shows the data structure of the request, Figure (b) shows the data structure of the information in the knowledge base 12 that is referenced by the AI ​​model 10, and Figure (c) shows the data structure of the response information from the generating AI server 100.

[0105] When a childcare worker wants to obtain health management information for one or more children, they input a request to generate health management information on the childcare worker terminal 210 or the like. When a request to generate health management information is input on the childcare worker terminal 210 or the like, a request including the input generation request and parameters is sent to the generation AI server 100. The request consists of, for example, the generation request "Please tell me about points to note regarding health management," and as parameters, a reference request and identification information "Nursery school: Midori no Oka Nursery School... Visual examination and parent comments," as shown in Figure 15(a). Of these parameters, "reference information" indicates that the AI ​​model 10 should refer to the information from the knowledge base 12 at the time of arrival, specifically (1) the child's identification information, (2) the child's name, (3) the child's body temperature, and (4) the visual examination and parent comments. Therefore, the information from the knowledge base 12 that the AI ​​model 10 refers to is configured as shown in Figure 15(b), for example.

[0106] In response to this request, health management information is obtained from the generating AI server 100. The health management information is structured as shown in Figure 15(c).

[0107] Health management information includes information regarding children's medication. Medication administration refers to, for example, when a child receives a prescription from a doctor for medication to be taken three times a day, and a caregiver gives the child one dose to be taken after lunch. Caregivers need to take meticulous care to avoid forgetting to administer the medication, not giving medication to other children, not giving the wrong medication, and not missing the timing of administration. However, since different children attend the daycare every day, it is difficult to keep track of each child's medication situation on any given day, making it difficult to perform appropriate health management. Therefore, by having the AI ​​model 10 generate information regarding the medication administration of one or more children, caregivers can keep track of each child's medication situation. For example, as shown in Figure 15(b), if the visual inspection / parent comment is "Administer medication after lunch (medication request form provided)," then as shown in Figure 15(c), information regarding the child's medication can be obtained such as "Ayaka-chan (medication after lunch): It is important to administer medication at the correct timing. After administering the medication, carefully monitor for any changes in her condition and keep a record."

[0108] [9. Dietary Management Information] Figure 16 shows the data structure when generating information on a child's dietary management. Figure (a) shows the data structure of the request, Figure (b) shows the data structure of the information in the knowledge base 12 that the AI ​​model 10 refers to, and Figure (c) shows the data structure of the response information from the generating AI server 100.

[0109] When a childcare worker wants to obtain meal management information for one or more children, they input a request to generate meal management information and image information of the meal menu at the childcare worker terminal 210, etc. When the childcare worker terminal 210, etc. receives the request to generate meal management information and image information, a request including the input generation request, parameters, and image information is sent to the generation AI server 100. The request consists of, for example, the generation request "Is there a risk for the child... Are there any risks?", the reference request and identification information "Nursery school: Midori no Oka Nursery School... Allergy information" as parameters, and image information (not shown), as shown in Figure 16(a). The "reference information" among these parameters is information from the knowledge base 12, specifically information from lunchtime, and indicates that the AI ​​model 10 should refer to (1) the child's identification information, (2) the child's name, (3) visual examination / parent comments, and (4) the child's allergy information. Therefore, the information from the knowledge base 12 that the AI ​​model 10 refers to is configured as shown in Figure 16(b), for example.

[0110] In response to this request, meal management information is obtained from the generating AI server 100. The meal management information is structured as shown in Figure 16(c).

[0111] [10. Generation of emergency response information] Figure 17 shows the data structure when generating emergency response information regarding children's health. Figure (a) shows the data structure of the request, Figure (b) shows the data structure of the information in the knowledge base 12 that is referenced by the AI ​​model 10, and Figure (c) shows the data structure of the response information from the generating AI server 100.

[0112] When a childcare worker wants to obtain emergency response information regarding the health of one or more children, they input a request to generate emergency response information on the childcare worker terminal 210, etc. When the childcare worker terminal 210, etc. receives a request to generate emergency response information, a request including the input generation request and parameters is sent to the generation AI server 100. The request consists of, for example, the generation request "Explain what happened based on the incident...please share it," and the parameters, a reference request and identification information "Childcare center: Midori no Oka Nursery School...Emergency response information regarding health," as shown in Figure 17(a). Of these parameters, "Reference Information 1" is information from the knowledge base 12 regarding an emergency that occurred during lunchtime, and indicates that the AI ​​model 10 should refer to (1) the child's identification information, (2) the child's name, (3) the details of the emergency, (4) the child's body temperature, (5) the visual examination and parent comments, (6) the child's allergy information, and (7) the vaccination history. Therefore, the information from the knowledge base 12 that the AI ​​model 10 refers to is configured as shown in Figure 17(b), for example. Furthermore, "Reference Information 2" indicates that the AI ​​model 10 will be given access to emergency response information related to health from the information in the knowledge base 12.

[0113] In response to this request, emergency response information is obtained from the generating AI server 100. The emergency response information is configured as shown in Figure 17(c).

[0114] Similarly, emergency response information can be obtained in the following forms: (1) emergency response information related to the environment, (2) emergency response information related to security, or (3) other emergency response information.

[0115] [Notification information generation process] Next, we will explain the case where the generation AI server 100 generates notification information using the AI ​​model 10.

[0116] Breaking news, signals, alerts, and other notifications are sent to the childcare worker terminal 210, etc., through one of the following configurations.

[0117] The first configuration inputs a predetermined prompt and necessary reference requests to the AI ​​model 10 when predetermined conditions are met. Here, necessary reference requests mean that if a reference request or a reference request and identification information is necessary, it is input, but if it is not necessary, it is not input. The same applies to configurations 2 to 4. The inference results of the AI ​​model 10 are then obtained, and notification information including the obtained inference results is sent to the childcare worker terminal 210, etc. As a prompt, for example, a prompt corresponding to the request exemplified in the response information generation process in Figure 6 (a prompt generated based on the generation request included in the request) can be adopted. As conditions, for example, (1) whether the current date is a predetermined date, time, or day of the week, (2) whether the place where childcare is to be performed, childcare is being performed, or childcare that has been completed (hereinafter simply referred to as "childcare" in this paragraph) is a predetermined place, (3) whether the childcare worker performing the childcare is a predetermined person or group, (4) whether the childcare is of predetermined content, and (5) whether the child's activity level, number of contacts, or BLE intensity is above or below a predetermined level.

[0118] The second configuration stores a table in the storage device 42 that associates conditions, prompts, and necessary reference requests. When a condition in the table is met, the corresponding prompt and necessary reference request are retrieved from the table and input into the AI ​​model 10. The AI ​​model 10 then retrieves its inference results and sends notification information, including the retrieved inference results, to the childcare worker terminal 210, etc.

[0119] The third configuration inputs predetermined prompts and necessary reference requests to the AI ​​model 10 and obtains the inference results from the AI ​​model 10. If the inference results meet predetermined conditions, notification information including the inference results is sent to the childcare worker terminal 210 or the like.

[0120] The fourth configuration involves storing a table in the storage device 42 that associates conditions, prompts, and necessary reference requests, retrieving prompts and necessary reference requests from the table, inputting the retrieved prompts and necessary reference requests into the AI ​​model 10, and obtaining the inference result of the AI ​​model 10. If the inference result satisfies the corresponding conditions in the table, notification information including the inference result is sent to the childcare worker terminal 210, etc.

[0121] Next, the effects of this embodiment will be described. In this embodiment, activity information regarding the number of times childcare workers interact with children, as well as activity information regarding the amount of activity and number of interactions with children, is registered in the knowledge base 12. A request including a request to generate information about children is input to the AI ​​model 10, and in response to the request, the AI ​​model 10 is made to refer to the activity information in the knowledge base 12 and the output information is obtained.

[0122] This allows us to obtain information based on the number of times childcare workers interact with children, as well as the amount of activity and interaction with children. As a result, we can understand information based on children's individuality and needs, compared to conventional methods.

[0123] Furthermore, in this embodiment, activity information for multiple children was registered in the knowledge base 12.

[0124] This allows us to obtain information based on the activity levels and frequency of contact of multiple children, enabling us to understand information based on each child's individuality when providing childcare for multiple children.

[0125] Furthermore, in this embodiment, the request to generate information about children is a request to generate information about the planning, evaluation, or improvement of group childcare.

[0126] This makes it possible to obtain information regarding the planning, evaluation, or improvement of integrated childcare, thus facilitating the planning, evaluation, or improvement of integrated childcare.

[0127] Furthermore, in this embodiment, the request to generate information about children is a request to generate evaluation information related to the evaluation of developmental policies.

[0128] This makes it possible to obtain evaluation information regarding the evaluation of policies or the evaluation of childcare workers on those policies, thus facilitating the evaluation of policies or the evaluation of childcare workers on those policies.

[0129] Furthermore, in this embodiment, the request to generate information about children is a request to generate safety management information related to the safety management of children.

[0130] This makes it easier to manage children's safety by providing them with safety information.

[0131] In this embodiment, step S106 corresponds to the registration means of Invention 1, step S204 corresponds to the input means of Invention 1, and step S206 corresponds to the acquisition means of Invention 1.

[0132] [Variation] In the above embodiment and its modifications, the AI ​​model 10 was made to refer to information in the knowledge base 12. However, the AI ​​model 10 is not limited to this, and any information that should be referenced in the knowledge base 12 (for example, activity information, basic information, environmental information, qualitative information, quantitative information, image information, or knowledge data such as childcare guidelines) can be included in the request. This makes it applicable even to configurations that do not have a knowledge base 12.

[0133] Furthermore, in the above embodiment and its modified form, the activity information is configured as information relating to activity level, number of contacts, and BLE intensity, but it is not limited to this, and can be configured as information relating to one or two of the activity level, number of contacts, and BLE intensity.

[0134] Furthermore, in the above embodiments and their modifications, the AI ​​model 10 was made to refer to activity information regarding the number of times childcare workers interacted with the children, as shown in Figures 7 and 9 to 14. However, it is not limited to these cases, and it is also possible to omit the reference to activity information regarding the number of times childcare workers interacted with the children. Also, in Figures 8 and 15 to 17, the AI ​​model 10 was not made to refer to activity information regarding the number of times childcare workers interacted with the children, but it is not limited to these cases, and it is possible to omit the reference to activity information regarding the number of times childcare workers interacted with the children.

[0135] Furthermore, in the above embodiments and their modifications, the AI ​​model 10 was not made to refer to activity information regarding the childcare worker's activity level, as shown in Figures 7 to 15. However, it is not limited to this, and the AI ​​model 10 can be made to refer to activity information regarding the childcare worker's activity level.

[0136] Furthermore, in the above embodiments and their modifications, the AI ​​model 10 was made to refer to activity information regarding the number of times a child was in contact with the AI, as in Figures 7, 8, and 11-14. However, it is not limited to these cases, and it is possible to not refer to activity information regarding the number of times a child was in contact with the AI. Also, in Figures 9, 10, and 15-17, the AI ​​model 10 was not made to refer to activity information regarding the number of times a child was in contact with the AI, but it is not limited to these cases, and it is possible to refer to activity information regarding the number of times a child was in contact with the AI.

[0137] Furthermore, in the above embodiments and their modifications, the AI ​​model 10 was made to refer to activity information regarding the child's activity level, as shown in Figures 7 to 14. However, it is not limited to this, and it is possible not to refer to activity information regarding the child's activity level. Also, as shown in Figures 15 to 17, the AI ​​model 10 was not made to refer to activity information regarding the child's activity level. However, it is not limited to this, and it is possible to refer to activity information regarding the child's activity level.

[0138] Furthermore, in the above embodiments and their modified forms, as shown in Figures 7 to 14, the AI ​​model 10 was made to refer to activity information regarding the amount of activity or the number of contacts acquired at two different time points or locations. However, it is not limited to this, and the AI ​​model 10 can also refer to activity information regarding the difference, sum, variance, standard error, standard deviation, standard score, mean, median, mode, kurtosis, skewness, minimum, maximum, and other statistical quantities of that amount of activity or the number of contacts. Moreover, it is not limited to two different time points or locations; the AI ​​model 10 can also refer to activity information regarding the amount of activity or the number of contacts acquired at one time point or location, or at three or more time points or locations.

[0139] Furthermore, in the above embodiment and its modified examples, information regarding the planning, evaluation, or improvement of developmental measures was generated as shown in Figure 13. However, the invention is not limited to this, and it is possible to generate information regarding the planning, evaluation, or improvement of developmental measures proposed or implemented by a childcare worker or group thereof. This makes it possible to plan, evaluate, or improve developmental measures proposed or implemented by a childcare worker or group thereof.

[0140] Furthermore, in the above embodiment and its modified examples, information regarding the planning, evaluation, or improvement of developmental measures was generated as shown in Figure 13. However, the invention is not limited to this, and information regarding the planning, evaluation, or improvement of children or groups utilizing developmental measures can also be generated. This makes it possible to plan, evaluate, or improve children or groups utilizing developmental measures.

[0141] Furthermore, in the above embodiments and their modifications, information about multiple children was generated as shown in Figures 7 to 10 and Figures 12 to 17, but the invention is not limited to this, and information about a single child can be generated. Also, as shown in Figure 11, information about a single child was generated, but the invention is not limited to this, and information about multiple children can be generated.

[0142] Furthermore, in the above embodiments and their modifications, information about one or more children was generated as shown in Figures 7 to 17. However, the invention is not limited to this, and in addition to this, information about one or more childcare workers involved in the group childcare, developmental measures, and other events shown in Figures 7 to 17 can also be generated.

[0143] Furthermore, although the configurations shown in Figures 15-17 were described as independent configurations in the above embodiments and their modified forms, they are not limited to these and can be additional configurations to the configurations shown in Figures 7-14.

[0144] Furthermore, in the above embodiment and its modified form, a request including a generation request and a reference request and identification information as parameters was sent, but the invention is not limited to this, and (1) a request including a generation request, or (2) a request including a generation request and a reference request as parameters can be sent. In other words, the AI ​​model 10 can be made to (1) perform inference without referring to the data in the knowledge base 12, or (2) perform inference by referring to all the data in the knowledge base 12.

[0145] Furthermore, in the above embodiment and its modifications, vector data is registered in the knowledge base 12 in association with identification information, but the invention is not limited to this, and vector data can be registered without being associated with identification information. Also, data in any format can be used, not just vector data.

[0146] Furthermore, in the above embodiment and its modified form, the distance between the caregiver and the child was measured by BLE intensity. However, the invention is not limited to this, and the wearable device 220 worn by the caregiver and the child can be equipped with a location information sensor such as GPS, and the distance between the caregiver and the child can be calculated based on the detection results of the location information sensor.

[0147] Furthermore, while a wearable device 220 was used in the above embodiment and its modifications, the invention is not limited to this, and any device that can acquire activity information can be used. For example, a multi-advertising BLE beacon equipped with an acceleration trigger can be used.

[0148] Furthermore, in the above embodiment and its modified form, the generating AI server 100 integrates the AI ​​model 10, the knowledge base 12, and the processing steps S100 to S106 and S200 to S208. However, it is not limited to this configuration, and some functions can be configured on separate servers or the like.

[0149] Furthermore, while the above embodiments and their modifications are implemented as a network system, they are not limited to this and can also be implemented as a single device or application.

[0150] Furthermore, while the above embodiments and their modifications have described their application to a network system consisting of the Internet 199, the invention is not limited to this, and may also be applied to, for example, a so-called intranet that communicates using the same method as the Internet 199. Of course, it is not limited to networks that communicate using the same method as the Internet 199, but can be applied to any network using any communication method.

[0151] Furthermore, in the above embodiments and their modifications, the process shown in the flowcharts of Figures 4 and 6 was described in the case of executing a program that is pre-stored in ROM 32. However, the invention is not limited to this, and the program describing these procedures may be read into RAM 34 from a storage medium in which the program is stored and then executed.

[0152] Furthermore, the above embodiments and their variations are mutually applicable. Furthermore, while the above embodiments and their modifications applied the present invention to cases where childcare workers provide care for children, the invention is not limited to these cases and can be applied to other cases without departing from the spirit of the invention. For example, the present invention can be applied when childcare is provided by caregivers other than childcare workers, or when care is provided to recipients other than children. [Explanation of symbols]

[0153] 100…Generating AI server, 10…AI model, 12…Knowledge base, 30…CPU, 32…ROM, 38…I / F, 39…Bus, 40…Input device, 42…Storage device, 44…Display device, 200…Nursery terminal, 210…Nursery teacher terminal, 220…Wearable device, 199…Internet

Claims

1. A registration means for registering attribute information regarding the attributes of the person being cared for and comment information from the caregiver or guardian regarding the person being cared for into a knowledge base that can be referenced by the AI ​​model, An input means for inputting a request to the AI ​​model that includes a request to generate information regarding the health management of the person being cared for, A childcare support system characterized by comprising an acquisition means for obtaining information output from the AI ​​model by referring to the attribute information and comment information of the knowledge base in response to the aforementioned request.

2. An input means for inputting a request into an AI model that includes attribute information relating to the attributes of the person being cared for and comment information from the caregiver or guardian regarding the person being cared for, and includes a request to generate information relating to the health management of the person being cared for. A childcare support system characterized by comprising an acquisition means for acquiring information relating to the health management of the person being cared for, based on the attribute information and comment information output from the AI ​​model, in response to the aforementioned request.

3. In either claim 1 or 2, A childcare support system characterized in that the request to generate information regarding the health management of the person being cared for is a request to generate information regarding the administration of medication to the person being cared for.

4. In claim 1, The registration means registers the allergy information of the person being cared for in a knowledge base that can be referenced by the AI ​​model. The input means inputs a request to the AI ​​model that includes a request to generate information regarding the care recipient's meal management, The acquisition means is characterized by acquiring information output from the AI ​​model by referring to the allergy information in the knowledge base in response to the request.

5. In claim 2, The aforementioned request includes allergy information of the person being cared for and a request to generate information regarding the dietary management of the person being cared for. The acquisition means is characterized by acquiring, in response to the request, information regarding the dietary management of the person being cared for, based on the allergy information, which is output from the AI ​​model.

6. In claim 1, The registration means registers emergency information regarding the details of the emergency situation involving the care recipient, allergy information of the care recipient, or vaccination history information of the care recipient into a knowledge base that can be accessed by the AI ​​model. The input means inputs a request to the AI ​​model that includes a request to generate information regarding emergency response for the person being cared for. The acquisition means is characterized by acquiring information output from the AI ​​model in response to the request by referring to emergency information, allergy information, or vaccination history information in the knowledge base.

7. In claim 2, The aforementioned request includes emergency information regarding the nature of the emergency situation involving the care recipient, allergy information of the care recipient, or vaccination history information of the care recipient, and includes a request to generate information regarding emergency response for the care recipient. The acquisition means is characterized by acquiring, in response to the request, information regarding the emergency response of the person being cared for, based on the emergency information, allergy information, or vaccination history information output from the AI ​​model.

8. A registration method for registering the allergy information of care recipients in a knowledge base that can be referenced by an AI model, An input means for inputting a request to the AI ​​model that includes a request to generate information regarding the meal management of the person being cared for, A childcare support system characterized by comprising an acquisition means for obtaining information output from the AI ​​model by referring to the allergy information in the knowledge base in response to the aforementioned request.

9. An input means for inputting a request into an AI model that includes allergy information of the person being cared for and a request to generate information regarding the dietary management of the person being cared for, A childcare support system characterized by comprising an acquisition means for acquiring information on the dietary management of the person being cared for, based on the allergy information, which is output from the AI ​​model in response to the aforementioned request.

10. A registration means for registering emergency information regarding the details of an emergency involving a care recipient, allergy information of the care recipient, or vaccination history information of the care recipient into a knowledge base accessible to the AI ​​model, An input means for inputting a request to the AI ​​model that includes a request to generate information regarding emergency response for the person being cared for, A childcare support system characterized by comprising an acquisition means for obtaining information output from the AI ​​model by referring to emergency information, allergy information, or vaccination history information in the knowledge base in response to the aforementioned request.

11. An input means for inputting a request into an AI model that includes a request to generate information regarding emergency response for the care recipient, including emergency information regarding the details of an emergency situation involving the care recipient, allergy information of the care recipient, or vaccination history information of the care recipient. A childcare support system characterized by comprising an acquisition means for acquiring information regarding emergency response to the childcare recipient based on the emergency information, allergy information, or vaccination history information output from the AI ​​model in response to the aforementioned request.

Images