JP7368034B1 - Information processing methods, programs, and information processing systems that support child consultation services - Google Patents

Abstract

In particular, an object of the present invention is to realize a method that can reduce the burden on field personnel by generating sentences using a trained language model. [Solution] An information processing method for supporting child counseling services provided by a server connected to a user terminal of a user related to child counseling via a network, wherein a control unit of the server At least child identification information that identifies the child, assessment information for determining whether to protect or support the child, and investigation record information obtained through the investigation are received as input information from the user terminal, and the input information is is stored in a storage unit of the server, and text data for instructing generation of a summary sentence based on at least the child identification information, the assessment information, and the investigation record information acquired from the storage unit is generated. [Selection diagram] Figure 18

Description

The present invention relates to an information processing method, program, and information processing system that support child consultation services.

Traditionally, various efforts have been made by various organizations to assess the risks associated with child abuse and protect children from abuse. However, issues such as a shortage of on-site personnel and an increase in the number of child abuse cases have been raised.

Against this background, as an automated technology for determining the risk of abuse, for example, Patent Document 1 discloses a technology for capturing an image of a child's body and automatically determining signs of abuse based on the captured image. There is.

Patent No. 6801902

Although automated technology such as image-based abuse symptom determination, as disclosed in Patent Document 1, can be expected to reduce the burden on on-site staff, there is a need for further measures to reduce the burden. In addition, such issues are not limited to information regarding child abuse between child guidance centers and various local government agencies, but also information related to public organizations such as municipalities and the national government (particularly information regarding domestic violence, sexual violence, and bullying). , information related to social issues such as harassment, crime, delinquency, physical disabilities, mental disorders, developmental disorders, child poverty, and elder abuse, areas related to food, clothing, shelter, and physical safety, and to provide peace of mind such as discrimination and abusive language. The same applies to information related to related areas, etc.).

Therefore, the present invention is a method that can reduce the burden on field personnel and support investigations, aid policies, and support plans by generating sentences using a trained language model. The purpose is to realize the following.

In one aspect of the present invention, there is provided an information processing method for supporting child counseling services provided by a server connected to a user terminal of a user related to child counseling via a network, wherein the control unit of the server is , accepting as input information at least child identification information for identifying the child, assessment information for determining at least one of protection or support for the child, and investigation record information obtained through the investigation from the user terminal; Storing the input information in a storage unit of the server, and generating text data for instructing generation of a summary sentence based on at least the child identification information, the assessment information, and the investigation record information acquired from the storage unit. do.

According to the present invention, in particular, by generating sentences using a trained language model, the burden on the person in charge at the site can be reduced.

1 is a block configuration diagram showing an information processing system according to a first embodiment of the present invention. 2 is a functional block configuration diagram showing the server 100 in FIG. 1. FIG. 2 is a functional block configuration diagram showing the user terminal 200 of FIG. 1. FIG. 3 is a diagram showing an example of child data stored in the server 100. FIG. 3 is a diagram showing an example of reception data stored in the server 100. FIG. FIG. 3 is a diagram showing an example of a check sheet (front surface) used when registering reception data. It is a figure which shows an example of the check sheet (back side) used when registering reception data. 3 is a diagram showing an example of user data stored in the server 100. FIG. 1 is an example of a flowchart related to an information processing method according to the first embodiment of the present invention. It is another example of the flowchart concerning the information processing method based on the first embodiment of the present invention. It is yet another example of a flowchart related to the information processing method according to the first embodiment of the present invention. FIG. 3 is a diagram illustrating an example of an application home screen displayed on a user terminal. FIG. 3 is a diagram showing an example of a screen displayed on a user terminal for requesting a risk assessment evaluation of an application. FIG. 3 is a diagram illustrating an example of a screen displayed on a user terminal for registering an image of an application. FIG. 7 is a diagram illustrating an example of a screen displayed on a user terminal for checking changes in risk assessment items of an application. FIG. 3 is a diagram showing an example of a screen displayed on a user terminal for using a chat communication function of an application. FIG. 3 is a diagram showing an example of a screen displayed on a user terminal for outputting a risk assessment evaluation result of an application. It is an example of the flowchart of the sentence generation process based on the information processing method based on 1st embodiment of this invention. 2 is a functional block configuration diagram showing the server 100 of FIG. 1 according to a second embodiment of the present invention. FIG.

Embodiments of the present invention will be described below with reference to the drawings. Note that the embodiments described below do not unduly limit the content of the present disclosure described in the claims. Furthermore, not all components shown in the embodiments are essential components of the present disclosure. Further, the features shown in each embodiment can be applied to other embodiments as long as they do not contradict each other.

(Embodiment 1)
<Configuration>
FIG. 1 is a block configuration diagram showing a business support system (information processing system) according to a first embodiment of the present invention. This system 1 consists of a user terminal 200 used by users related to child consultation such as child consultation center staff, a child consultation record system 300 that records/manages information related to child consultation, and a learning system based on input text data. The system is composed of a text generation system 400 that generates sentences using an established language model, and a server 100 that mediates between the two. Note that all or part of the functions of the child consultation recording system 300 can be provided in the server 100, and if all the functions are provided in the server 100, even if the child consultation recording system 300 is unnecessary in this system 1. good. Alternatively, all or part of the functions of the child consultation recording system 300 may be implemented as functions of the user terminal 200 by an application executed on the user terminal 200. Similarly, all or part of the functions of the text generation system 400 can be provided in the server 100, and if all the functions are provided in the server 100, the text generation system 400 may not be necessary in the present system 1. Alternatively, all or part of the functions of the text generation system 400 may be implemented as functions of the user terminal 200 by an application executed on the user terminal 200.

Server 100 and user terminal 200 are connected via network NW. The network NW includes the Internet, an intranet, a wireless LAN (Local Area Network), a WAN (Wide Area Network), and the like.

The server 100 provides an application for supporting business efficiency to a plurality of user terminals 200A, 200B related to users related to child consultation such as staff of a child guidance center, and provides an application for supporting work efficiency to a plurality of user terminals 200A, It is a device that manages information registered from each of the 200B and provides evaluation results and research findings based on the registered information, such as risk assessment evaluation regarding the possibility of child abuse. , a general-purpose computer such as a workstation or a personal computer, or may be logically implemented by cloud computing. In this embodiment, one server is exemplified for convenience of explanation, but it is not limited to this, and there may be multiple servers, or servers with different roles such as an authentication server and a database server. good.

As mentioned above, the user terminals 200A and 200B are, for example, personal computers or tablets owned by users related to child counseling such as staff of child guidance centers and who use the application provided by the server 100. Although the information processing device is a terminal or the like, it may be configured by a smartphone, a mobile phone, a PDA, or the like. Hereinafter, for convenience of explanation, the user terminals 200A and 200B will be collectively referred to as the user terminal 200.

In addition, this system is connected to the child consultation recording system 300 via the network NW, and the user using the user terminal 200 can send child information related to the child to the server 100 or the child consultation recording system 300. The server 100 can send and receive information to and from the child consultation recording system 300, and can coordinate data. Here, the child consultation record system 300 is a system whose main function is to issue records of child consultations and administrative documents. This system issues notifications, manages family information and contributions in conjunction with administrative information, and manages the progress of procedures. On the other hand, this system is a system whose main function is to support communication and decision-making with users, and it allows input, viewing, and sharing of records inside and outside the child guidance center, and allows chatting between child guidance center staff members. This system performs communication, registration and sharing of children's photos, and simulation of past temporary protection trends. As mentioned above, the server 100 and the child consultation record system 300 are connected via the network NW, and data is linked with each other using, for example, a QR code (registered trademark), child number, or reception number as a key, so There is no need for the person to input and register information in both the server 100 and the child consultation record system 300, and by registering information in one system, it is possible to synchronize with the other system. As mentioned above, if the server is equipped with all or part of the functions of the child consultation record system 300 and the processing can be completed on the server 100, data linkage using a QR code (registered trademark) etc. is not necessary. It may also be a thing.

In addition, this system is connected to the text generation system 400 via the network NW, and a user using the user terminal 200 can transfer text data to the learned language model of the text generation system 400 via the user terminal 200. can be input to generate sentence data corresponding to the text data. Here, the text generation system 400 receives text data such as questions and requests in a chat format using a large-scale language model such as ChatGPT, and receives text data such as a learned language model (for example, a large-scale language model). ) and generate data in response to the question text or text according to the request (for example, a summary text based on a summary request and original text, an example text based on an example request, etc.). The trained language model may be, for example, a learning model that has been trained using a large amount of data on the Web as learning data, and in particular, the learning model may be a learning model provided from an external source. The learning model may be a learning model in which a user or a server provider (namely, an administrator) performs additional learning on a specialized field such as a child consultation field.

In this embodiment, the system 1 will be described as a configuration in which each user uses a user terminal 200 to operate the server 100. However, the server 100 is configured as a standalone, and the server itself has It may also be provided with a function for performing operations.

FIG. 2 is a functional block configuration diagram of the server 100 in FIG. 1. The server 100 includes a communication section 110, a storage section 120, and a control section 130.

The communication unit 110 is a communication interface for communicating with the user terminal 200, the child consultation recording system 300, and the text generation system 400 via the network NW, and includes, for example, TCP/IP (Transmission Control Protocol/Internet Protocol), closed area Communication is performed according to the communication rules of the network, etc.

The storage unit 120 stores programs, input data, etc. for executing various control processes and functions within the control unit 130, and is composed of RAM (Random Access Memory), ROM (Read Only Memory), etc. Ru. The storage unit 120 also includes a child data storage unit 121 that stores various data related to users, a child data storage unit 121 that stores various data related to children, a reception data storage unit 122 that stores various data related to reception; It has a user data storage section 123 and the like that stores various data related to users. Furthermore, the storage unit 120 can also temporarily store data communicated with the user terminal 200, the child consultation recording system 300, and the text generation system 400. Note that a database (not shown) storing various data may be constructed outside the storage unit 120 or the server 100.

The control unit 130 controls the overall operation of the server 100 by executing programs stored in the storage unit 120, and is composed of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), etc. Ru. The functions of the control unit 130 include a reception unit 131 that receives input from the user terminal 200 or the child consultation record system 300, a data management unit 132 that refers to and processes various data, and a data management unit 132 that refers to and processes information input by the user. It includes a risk evaluation unit 133 that performs analysis and evaluation, a text generation instruction unit 134, and the like. The reception unit 131, data management unit 132, risk evaluation unit 133, and text generation instruction unit 134 are activated by a program stored in the storage unit 120 and executed by the server 100, which is a computer (electronic computer).

The reception unit 131 is provided by the server 100, and when the user makes a predetermined input via a user interface such as a screen displayed via a web browser or an application on the user terminal 200, Instructions and various information are received from the terminal 200 via the communication unit 110. Alternatively, information is received from the child consultation recording system 300 via the communication unit 110 in response to a request from the server 100 or the user terminal 200. Alternatively, text data for instructing text generation (also referred to as "text generation instruction information") is sent from the server 100 or user terminal 200 to the text generation system 400, and the text generation system 400 generates the text according to the text data. The written text data is received via the communication unit 110.

The data management unit 132 manages various data related to users, such as various user data, child data, reception data, evaluation data, etc., and performs predetermined procedures such as registration and reading into the corresponding data storage unit in the storage unit 120. Process.

The risk evaluation unit 133 performs processing to evaluate the risk of child abuse based on information registered from the user terminal 200 (or the child consultation recording system 300).

The text generation instruction unit 134 instructs the generation of a text based on at least the child identification information acquired from the storage unit 120 (for example, the reception data storage unit 122), risk assessment information (described later), and investigation record information (described later). The text data (text generation instruction information, including so-called prompts) is generated. The generated text data is stored in the storage unit 120 in association with at least child identification information (furthermore, in association with input date and time information, reception ID, etc.). The text instructing the above generation may be, for example, a summary text of risk assessment information and investigation record information linked to the child, various diagnostic documents and administrative documents with designated administrative formats, and documents submitted to judicial institutions. Automatic generation is possible. In addition, the text instructing generation may be, for example, user support information estimated from risk assessment information and investigation record information at one or more points in the past linked to child information (in particular, what points should be investigated, what points This could be information such as advice such as whether to consider Note that the text data for instructing the generation of sentences is not limited to child identification information, risk assessment information, and investigation record information; any of the information may be deleted, or the information described later may be The reception data may include information on any one of the reception data.

FIG. 3 is a functional block diagram showing the user terminal 200 of FIG. 1. As shown in FIG. The user terminal 200 includes a communication section 210, a display operation section 220, a storage section 230, a camera 240, and a control section 250.

The communication unit 210 is a communication interface for communicating with the server 100 via the network NW, and communication is performed according to a communication protocol such as TCP/IP.

The display operation unit 220 is a user interface used by the user to input instructions and display text, images, etc. in accordance with input data from the control unit 250. If the user terminal 200 is a smartphone or a tablet terminal, the user terminal 200 is composed of a touch panel or the like. This display operation section 220 is activated by a control program stored in a storage section 230 and executed by the user terminal 200 which is a computer (electronic computer). Via the display operation section, users can respond to the provided aptitude test by pressing the keyboard, moving the cursor using the mouse, or tapping, swiping, or pinching the touch panel. Operations, etc. can be performed.

The storage unit 230 stores programs, input data, etc. for executing various control processes and functions within the control unit 250, and is composed of a RAM, a ROM, and the like. Furthermore, the storage unit 230 temporarily stores the contents of communication with the server 100.

The camera 240 has, for example, a function of capturing an image of a child's body part.

The control unit 250 controls the overall operation of the user terminal 200 by executing a program stored in the storage unit 230, and is composed of a CPU, a GPU, and the like.

Note that the server 100 may be configured to include the function of a display operation section, or in this case, the server 100 may not be configured to include the user terminal 200.

Note that the functional configuration of the child consultation recording system 300 is also substantially the same as that of the server 100 or the user terminal 200, so a description thereof will be omitted.

FIG. 4 is a diagram showing an example of child data stored in the server 100.

Child data 1000 shown in FIG. 4 stores various data related to children. In FIG. 4, for convenience of explanation, an example of one child (child identified by child ID "10001") is shown, but information on multiple children can be stored. Various data related to children include, for example, child information regarding children (child's name, address, contact information such as email address, gender, age, school name, grade, homeroom teacher name, school attendance status, tag information, etc.), protection personal information (parent's name, ID, address, contact information, gender, etc.), related person information (related person's name, ID, address, contact information, gender, etc.), family group information (family group name, ID, group Name explanation, group members (including guardians due to past marital status and common-law relationships), information on related institutions (related institution name, ID, type (medical institution, police, educational institution, etc.), address, contact information) ), etc.

FIG. 5 is a diagram showing an example of reception data stored in the server 100.

The reception data 2000 shown in FIG. 5 stores various data related to reception. In FIG. 5, for convenience of explanation, an example of one reception (reception identified by reception ID "20001") is shown, but information on multiple receptions can be stored. Normally, the reception desk may be registered in association with child data (e.g. child ID) related to the registered child, but depending on the management purpose, it may be necessary to It may be registered in association with any arbitrary ID such as a related institution ID) or a user ID (including a user institution ID related to the user institution to which the user belongs). In particular, in an emergency, the reception process may be performed before child data is registered, so in that case, the reception data should be managed in association with the user ID (user institution ID). , after registering the child data, the reception data may be registered in association with the child data (child ID). The reception data is data received by the reception unit 131, but after being associated with child information etc., it is managed as part of the associated information (for example, part of the child information). Good too.

As various data related to reception, for example, basic reception information (reception date and time, input person, type of abuse (physical abuse, neglect, etc.), response start date, etc.), emergency dispatch or investigation protection, etc. Risk assessment information related to childcare consideration, information regarding whether or not the child falls under limit assessment (items such as recurrent conditions that make the child's safety questionable, interview canceled, and the child is asked to stay home even when visiting the child), and information regarding the child. Image information, information on the child's wounds, information on the child's psychological diagnosis, information on the child's home situation, decision-making results and/or safety confirmation results, items related to corresponding implementation (temporary protection was implemented or temporary protection was lifted) whether measures have been taken against children's homes, etc., whether measures have been lifted against children's homes, etc., whether measures have been taken against foster parents, whether measures against foster parents have been lifted, and whether measures have been taken against foster parents, etc., and whether measures have been taken against foster parents, etc. whether social resources and services were implemented, whether a monitoring system was established or used in the local government or local community), who approved the survey record, and survey record information (survey date and time, survey title). , minutes of the meeting (investigation results (facts)), information on estimates, information on future responses, additional information obtained through the investigation (risk assessment information (information obtained through the investigation), images of children obtained during the investigation, investigation Child's wound information obtained at the time of the investigation, psychological diagnosis information of the child obtained at the time of the investigation, information regarding the child's home situation obtained at the time of the investigation, date and time of the next interview for the investigation, status in the investigation completion dictionary, safety at the time of the investigation completion. Scale (input result information in intuitive numerical values regarding child safety), response details, decision maker of investigation records, etc.), evaluation information (regarding past protection rate, probability of falling under serious abuse, recurrence rate, response) number of days, information about investigations that need to be conducted, past similar cases (information about the applicable abuse type among multiple abuse types), graphs (e.g., radar charts (probability of serious abuse, recurrence rate, number of response days) etc.), a bar graph, a bubble chart, etc., and may be a graph for each reception date and time (input date and time), or a graph that can be compared in chronological order according to the reception date and time (input date and time). )), recommendation information (recommendations and reference information regarding future actions), etc.In addition, at the time of reception, data regarding at least part of the various data exemplified above is accepted.

Risk assessment information includes, for example, input date and time information, serious confirmation item check information (for example, S rank items shown in Figure 6, and if even one is checked, it is determined that safety should be ensured. (including at least either item check information for considering emergency dispatch or item check information for considering investigation protection), check information for items requiring investigation (for example, as shown in FIG. 7). Includes check information regarding A and B rank items that are judged to require safety consideration if even one is checked.In particular, check information regarding items to be confirmed during the investigation) .

A limit assessment is a promise made with a parent or guardian, in which a promise is made in advance to force intervention if the promise is broken. Despite this, items are included to confirm whether there are situations in which the child's safety cannot be confirmed, such as canceling a designated interview without permission, refusing to visit the child's home, or if the child has an injury. The information regarding whether or not the limit assessment applies may be information indicating whether or not one or more of the above-described limit assessment items apply. Items related to limit assessment may be set in common, or may be arbitrary items set by the user in association with at least one of child information and guardian information.

6 and 7 are diagrams showing an example of a check sheet used when registering reception data.

The check sheet includes a checklist as shown in FIGS. 6 and 7 as an example. The checklist includes serious items (S-rank items) to be checked before determining whether temporary protection (notification/referral) is necessary, A-rank items that need to be investigated before determining whether temporary protection (notification/referral) is necessary, and assistance policy. It is broadly divided into three categories: B-rank items that need to be investigated before the meeting, and S-rank items that include items related to severe physical abuse, items related to serious neglect, items related to suspected sexual abuse, and items related to serious Other items related to abuse are divided into medium categories, and the medium categories are ultimately divided into 10 to 20 check items. For each check item, for example, child guidance center staff will check three things: suspicions before the investigation, suspicions after the investigation, and confirmation of the situation through the investigation. Further, the check sheet includes, for example, a column in which a child guidance center staff member or the like records decisions regarding temporary protection of children and the reasons for the decisions. In addition, the check sheet is provided with, for example, a column for entering information corresponding to the reception data.

FIG. 8 is a diagram showing an example of user data stored in the server 100.

User data 3000 shown in FIG. 8 stores various data related to users. In FIG. 4, for convenience of explanation, an example of one user (a user identified by user ID "30001") is shown, but information on multiple users can be stored. Various data related to users include, for example, basic information about users (user name, address, contact information such as email address or telephone number, affiliation information, position information, access authority information, etc.), Contains related child information (child ID corresponding to or corresponding to the user, child ID linked to accepted child information, child ID linked to provided child information, etc., especially list information), etc. be able to. In addition, the user data may have a user ID (account) registered for each person belonging to the user institution, such as a staff member, but instead of or in addition to this, a common user ID for each user institution may be registered. (account) may be registered.

<Processing flow>
The flow of processing of the information processing method executed by the information processing system 1 of this embodiment will be described with reference to FIG. 9. 1 is an example of a flowchart related to an information processing method according to the first embodiment of the present invention.

Here, in order to use this system 1, a user (for example, a user related to child counseling such as a staff member of a child guidance center) uses each web browser or application on the user terminal 200. When accessing the server 100 and using the service for the first time, the user registers basic information in the user data 3000 for new user registration and obtains a user account. If you have already acquired a user account, you can use the service by logging in after receiving predetermined authentication, such as entering your ID and password. After this authentication, a predetermined user interface screen is provided via a website, application, etc., and the process proceeds to step S101 shown in FIG. Here, FIG. 12 shows an example of an application home screen displayed on the user terminal 200. The user can perform the registration described below by selecting the registration menu on this screen. In FIG. 12, list information of registered child data, corresponding status information, update date information, evaluation information, payer information, etc. are shown.

First, as processing in step S101, the reception unit 131 of the control unit 130 of the server 100 receives registration of basic information such as child information regarding a child from the user terminal 200 via the communication unit 110. The data management unit 132 of the control unit 130 of the server 100 stores the received basic information in the child data storage unit 121 of the storage unit 120 as child data 1000 in association with the child ID. Here, if the user has registered the above basic information in the child consultation record system 300, the user reads the QR code (registered trademark) with the camera built into the user terminal 200 and displays the child number on the displayed screen. Basic information registered in the child consultation recording system 300 can also be imported into the application provided by the server 100 by inputting a number such as .

Here, as shown in FIG. 10, as basic information, the reception unit 131 collects, in no particular order, the child's name, address, contact information such as e-mail address, gender, age, school name, grade, homeroom teacher name, Registration of child information regarding the child, such as school attendance status and tag information, is accepted (step S201), and registration of guardian information regarding the guardian, such as the guardian's name, ID, address, contact information, gender, etc., is accepted (step S202). ), then we accept registration of related party information regarding the child's related persons, such as the name, ID, address, contact information, gender, etc. of related persons such as uncles, aunts, friends of the child's older brother, and dating partner of the child's mother ( Step S203), registration of family group information regarding the family group, such as family group name, ID, group name description, group members (including children, siblings, guardians from past marriages, common-law relationships, etc.) (Step S204), and also accepts registration of related organization information regarding related organizations, such as related organization name, ID, type (medical institution, police, educational institution, etc.), address, contact information, etc. (Step S205). can.

Next, as processing in step S102, the reception unit 131 receives registration of reception information from the user terminal 200. First, the user inputs reception data such as basic reception information (type of abuse (physical abuse, neglect, etc.), start date and time of response, etc.) on the application screen displayed on the user terminal 200. The data management unit 132 of the control unit 130 of the server 100 stores the received reception data in the reception data storage unit 122 of the storage unit 120 as reception data 2000 in association with the reception ID. Here, if the user has registered the above-mentioned reception information in the child consultation record system 300, the user performs a predetermined reading and registration operation on the user terminal 200 (for example, using the camera built in the user terminal 200 to register the QR code). trademark), inputting a number such as a child number on the displayed screen, or sending a request to read child data from the server 100 or user terminal 200 to the child consultation record system 300. It is also possible to import reception data registered in the child consultation record system 300 into an application provided by the server 100.

As shown in FIG. 11, in relation to the reception information, the reception unit 131 receives information from the user terminal 200, including the response date and time, investigation title, minutes (investigation results (facts)), estimates, and future actions. (step S301), and registration of child image-related information such as child image information and/or child wound information, child psychological diagnosis information, and information regarding the child's home situation. Acceptance (step S302), then acceptance of registration of risk assessment information (information obtained before the investigation or information obtained through the investigation) (step S303), and communication between multiple users via the chat application. (step S304), and generates a sentence using the sentence generation system 400 based on the child identification information, risk assessment information, investigation record information, etc. acquired from the storage unit 120 (step S304). Step S305). The data management unit 132 of the control unit 130 of the server 100 stores the received basic information in the reception data storage unit 122 of the storage unit 120 as reception data 2000 in association with the reception ID.

Here, the registration of the above-mentioned investigation record information is assumed to be an operation performed by the staff within the child guidance center based on the information obtained from the staff dispatched outside the institution, and the registration of the above-mentioned child image-related information and the above-mentioned risk assessment information. Registration is assumed to be an operation carried out by staff dispatched outside the school to confirm the safety of children, and regarding the chat communication mentioned above, staff dispatched outside the school will report the results of the investigation and report the results within the school. A series of registered information may be input and accepted by multiple staff members inside and outside the institution, such as an operation to confirm the response method sent by a staff member.

Here, FIG. 14 shows an example of an image registration screen. On the image registration screen, the user can upload an image (for example, an image taken with a camera built into the user terminal 200 or an image saved on the user terminal 200 after being taken with a mobile device such as a digital camera or smartphone). Select wound, psychological diagnosis, home situation, etc. as the type of image to be imported. If wound is selected as the image type, select the part of the child's body related to the wound, and select the type of wound/fracture such as scar, bruise, bite mark, burn, hair loss, bloodshot eyes, etc. You can then select the details of the wound/fracture from a list (discoloration, bleeding, swelling, sore, old injury, keloid, multiple or mixed, deformation/enlargement/deficiency, etc.). Images related to psychological diagnosis may be photographed images of pictures drawn by children, images of developmental characteristic findings received at medical institutions, and the like. Images related to the condition of the house include the condition in the room (especially the garbage situation, etc.) and things in the house that caused the injury (for example, if there is evidence of a fall down the stairs, a photographed image of the stairs of the house, etc.) ) etc. In addition, any images related to children may be used, including but not limited to images where the object that caused the wound is outside the home (e.g., at a park or school). may be separately selected and saved as images related to related facilities.

Further, FIG. 16 shows an example of a chat communication screen. By selecting the icon for using chat communication displayed at the top of the application screen, the user will be presented with a list of individual or group names with whom to communicate via chat. By selecting this option, users can communicate via chat.

Here, in order to input risk assessment information, the user selects the risk assessment information recording menu on the reception information input screen as shown in FIG. 13, and selects the risk assessment information registration form. be able to. Specific examples of the information accepted in the risk assessment information registration form (various data related to the above-mentioned reception) include: 1) date and time of reception (input the date the risk assessment was conducted, etc.); 2) input person (input person) 3) Information at the time of notification reception (enter and select the information at the time of notification reception), 4) Emergency dispatch (select the appropriate risk assessment item to consider emergency dispatch), 5) Investigation and protection consideration (Select risk assessment items to consider investigation protection), 6) Limit assessment applicable (Check whether it falls under limit assessment), 7) Attach images (“Wound”, “Psychological diagnosis”, “Home 8) Decision-making, safety confirmation (select the result of decision-making. Based on the input information, You can check simulation results such as recurrence rate, probability of serious abuse, number of response days, etc., and past similar cases), 9) Action items (choose the actions taken and enter the response date and time), 10) Approval person (Set the person who approves the investigation record) etc. Further, the registration form may have a function of copying at least part of the contents registered in the past (such as selecting a child ID related to risk assessment information to be copied and generating a copy). In addition, as shown in the example screen in Figure 15, by selecting the "Risk Assessment History" menu, the user can view at least some of the items included in the risk assessment information (for example, at least one of the items for emergency dispatch or investigation protection). The risk assessment information (items related to considering whether to do one or the other) is displayed so that the updated information for each item of risk assessment information for each child can be confirmed in chronological order (i.e., changes can be confirmed) based on at least date and time information. do. This can be done, for example, by registering the child in association with the date and time information and the child's ID. In inputting risk assessment information, the user may refer to the check sheets illustrated in FIGS. 6 and 7 and accept input operations as assessment items via the application on the user terminal 200, or the check sheet and at least A user interface screen that allows partially identical or similar items to be entered as an input form on the application may be displayed to accept input operations, or a check sheet image taken by the camera 240 of the user terminal 200 may be image-analyzed. The check details may be entered automatically. Furthermore, it may be possible to change the contents of at least some of the assessment items (for example, mark them as checked or uncheck them) on the screen on which the transition of the risk assessment information is displayed.

Next, returning to step S103 in FIG. 9, the risk evaluation unit 133 of the control unit 130 performs a risk assessment regarding the possibility of child abuse based on the basic information of the registered child and reception data (particularly risk assessment information). Perform an evaluation (i.e., generate evaluation data). The generated evaluation data may be stored as reception data in association with the reception ID. For example, the risk evaluation unit 133 receives, as input information, past basic information of the child (age, gender, address (municipality)), reception information (type of abuse, reception hours, reception classification (new, re-notification, re-application), and Risk assessment information associated with reception information (reception route, main abuser, status of application to each item of risk assessment), information regarding whether or not limit assessment is applicable, image information about the child, information on child's wounds, child's psychology Diagnosis information, information on the situation of the child's home, decision-making result information and/or safety confirmation result information, items related to corresponding implementation (whether temporary protection was implemented, whether temporary protection was lifted, measures taken at child care facilities, etc.) measures taken against child care facilities, etc., measures taken against foster parents, measures taken against foster parents, cooperation with related organizations, social resources and services. information on whether the local government or local community has constructed or used a monitoring system), information on mitate, information on future responses, and survey record information (survey date and time, survey title, minutes). (Investigation results (facts), additional information obtained through the investigation (risk assessment information (information obtained through the investigation), images of children obtained during the investigation, information on children's wounds obtained during the investigation, Psychological diagnosis information and information on the child's home situation obtained during the investigation, date and time of the next interview for the investigation, status at the time of completion of the investigation, safety scale at the time of completion of the investigation (input of intuitive numerical values regarding child safety) If you enter some or all of the information (result information), response details, etc.), the output information will be the past protection rate (an indicator of the tendency to be temporarily protected), past similar cases (classification of abuse into multiple types), etc. Displays the characteristics of the type that the current case is most similar to. If there are past cases belonging to the same type, displays the past case records), as well as the probability of serious abuse, recurrence rate, and number of response days. , a risk assessment learning model that outputs part or all of information such as the preventive effects of intervention or support (for example, it may be a learning model created by learning for each indicator, or it may be a common learning model). ), growth prediction models (e.g. height or weight measurement information of a predetermined number of people, measurement timing information (e.g. number of days elapsed since birth), basic information of the person to be measured (e.g. age (age) A learning model that shows growth prediction learned based on the following information (including gender, address, nationality, presence or absence of disease, presence or absence of abuse), and for each basic information item (i.e. age It may be a learning model created by learning for each gender, gender, etc.), or it may be a common learning model. ) may be used to generate evaluation data (output information). The learning model may be learned in advance as training data by extracting at least part of the above-mentioned input information and result information corresponding to the output information from past data, and is stored in the storage unit 120. or may be stored in an external server. In the teacher data during learning, if more types of input information are included among the above-mentioned input information, more precise output information (probability of falling under serious abuse, etc.) can be output. As another example, the risk evaluation unit 133 extracts similar past cases according to a combination of some or all of the input information (particularly risk assessment information), and extracts the results of the extracted past cases. Some or all of the above output information may be generated by calculating statistical values from the above. Note that the methods for generating these various types of information and evaluation data are merely examples, and are not limited to these. Further, the risk evaluation unit 133 may generate evaluation data by further referring to at least one of survey record information and photographic information, for example, as reception data to be referred to for evaluation.

Next, as processing in step S104, the risk evaluation unit 133 transmits the output information generated in the above step to the user terminal 200, and displays the output information in a predetermined format on the user interface screen of the application on the user terminal 200. Is displayed.

FIG. 17 is a diagram showing an example of a screen displayed on a user terminal for outputting a risk assessment evaluation result of a business support application. By selecting the "Simulation" tab on the personal page screen of the application displayed on the user terminal 200, the user can view information on past protection rates, probability of recurrence, number of response days, severity of illness, etc., as shown in FIG. A radar chart showing the characteristics of the case using three indicators related to severity (probability of serious abuse) and a screen visualizing past similar cases are displayed. Here, regarding information on past protection rates, if the value is high, it indicates that similar cases tended to be temporarily protected in the past, and from the perspective of whether or not to provide temporary protection, it is important for users to , to support judgment of urgency. In addition, the radar chart shows the severity (probability of falling into a classification defined as a serious case (e.g., 24 patterns) out of multiple abuse classifications (e.g., 36 patterns)) and recurrence rate (group of similar cases that apply). It represents the balance between the probability of recurrence (probability of recurrence) and the number of days required to complete the case (the number of days required to complete the case for the corresponding case group).If the area of the radar chart is large, it indicates that the disease is serious or the recurrence rate is high. , meaning that it tended to take a long time to complete. For past similar cases, past cases are classified into multiple types (for example, 36 types), and the characteristics of the pattern to which the case is most similar are displayed. If there are past cases that belong to the same pattern, the past case records are displayed. In this example, "case 47 of physical abuse" is displayed as case 47, and most of these cases were reported to the municipality, 30% of cases were injuries to the craniofacial or abdominal area, and 27% were signs of continued abuse. However, in 7% of cases, the whistleblower witnessed the violence and found it difficult to discuss it with their parents, a pattern seen in 5.2% of past cases. By checking past case records, users can get hints about how cases will change. In addition, it is not limited to the content shown in the example, but it is also possible to generate and present evaluation data, such as to what degree (or what kind) of serious abuse is involved, based on the applicable situation of the serious item. That's fine, and even if recommendations are presented, such as where to start the investigation, what kind of support should be provided, and whether it's best to collaborate with related organizations, based on information on the response content in similar cases. good.

The flow of text generation processing in the information processing method executed by the information processing system 1 of this embodiment will be described with reference to FIG. 18. First, the text generation instruction unit 134 instructs the generation of a text based on the child identification information acquired from the storage unit 120 (for example, the reception data storage unit 122) and reception information such as risk assessment information and investigation record information. Text data (text generation instruction information) is generated (step S401). One specific example of text data is, for example, ``Please provide a summary of the information in (1) to (4) below. (1) Child information (e.g., child identification information), (2) Reception information. (reception identification information), (3) Risk assessment item information (selection information such as applicable, suspected, not applicable, unknown, etc.), (4) Investigation record information (for example, the child's statement ``I was hit by my mother's boyfriend.''・Mother ``(silent)'' ・Common-law husband ``I don't know'', etc.)''. Another specific example of text data is, for example, ``The record of the first interview is ``...''.・The record for X + 1st time is "...".・The record for X+2nd time is "...". The prompt could be something like, ``Based on this information, what points should I investigate or what should I consider?'' Next, the text generation instruction unit 134 transmits the generated text data (text generation instruction information) to the text generation system 400 via the communication unit 110 (step S402). Next, the reception unit 131 receives the text data created by the text generation system 400 via the communication unit 110, and links it to at least the child identification information (and further links it to input date and time information, reception ID, etc.) It is stored in the storage unit 120 (step S403). Next, the data management unit 132 transmits output information including the received text data to the user terminal 200, and the output information is displayed in a predetermined format on the user interface screen of the application on the user terminal 200.

Note that the control unit 130 may further include a modification unit that modifies the text data received from the text generation system 400 via the user terminal 200, and links the text data modified by the modification unit to child information. The information may also be stored in the storage unit 120.

The control unit 130 also controls a predetermined format (for example, a document format for administrative submission, which is the format of various diagnostic documents and administrative documents specified by the administration, and the format of documents to be submitted to judicial institutions (for example, explanatory materials, etc.). etc.) from the storage unit 120, and inputs the text data received from the text generation system 400 via the user terminal 200 into the corresponding part of the format data as input data ), it becomes possible to automatically generate various document data (for example, document data for administrative submission). The generated document data may be stored in the storage unit 120 in association with at least child identification information (furthermore, in association with input date and time information, reception ID, etc.). By transmitting or printing the document data via a network, the burden of document submission is reduced.

(Embodiment 2)
<Configuration>
FIG. 19 is a functional block configuration diagram of a server 100 provided in a business support system (information processing system) according to the second embodiment of the present invention. Hereinafter, detailed description of the configurations similar to those of the first embodiment will be omitted, and it will be assumed that they have the same or similar functions.

The control unit 130 in the second embodiment further includes a needs evaluation unit 135. The needs evaluation section 135 may be provided together with the risk evaluation section 133 or in place of the risk evaluation section 133. Note that hereinafter, the risk evaluation section 133 and the needs evaluation section 135 may be collectively referred to as the "evaluation section."

In the first embodiment, risk assessment information is accepted, but in the second embodiment, needs assessment information is accepted instead of or in addition to risk assessment information. Like the risk assessment information, the needs assessment information is stored in the storage unit 120 in association with at least child identification information (furthermore, in association with input date and time information, reception ID, etc.). Note that hereinafter, risk assessment information and needs assessment information may be collectively referred to as "assessment information."

As mentioned above, risk assessment information is information for considering or making decisions about emergency dispatch and protection (for example, temporary protection, investigation protection, etc.). On the other hand, needs assessment information, like risk assessment information, is information that is linked to the child, but it is also information about the support that the child or at least someone related to the child (such as a family member or a member of a related organization) requests. It is. Support is the act of helping others solve problems that are difficult for the target person to solve by themselves, such as counseling support, living support, child rearing support, and financial support (public assistance). , employment support, school attendance support, medical support, psychological support, and support that bridges the organizations that provide these supports and the people concerned. This may include providing support and developing strengths that allow them to cope well with matters related to child safety. Note that the subject of risk assessment information is not limited to emergency dispatch and protection, but may also cover interventions in general. Interventions include interfering actions from others that aim at specific changes and transformations in the target person, such as (1) various types of actions such as referral to a child guidance center or local government, temporary protection, visits and interviews, etc. Separation of children and guardians (caregivers), or children and related persons other than guardians, such as investigation, placement in child care facilities, arrest or detention of abusers, etc. (2) For example, lifestyle guidance (3) Physical and mental issues faced by the individual child or guardian (caregiver) (e.g., implementation of medical measures (e.g., prescription of drugs) and psychological therapy to improve or alleviate diseases (diseases, disabilities, etc.), (4) welfare assistance for poverty and job loss, and meals at children's cafeterias. It can be support, etc.

Needs assessment information includes, for example, input date and time information, needs item check information (check items for estimating the support required by at least one of the child or a person related to the child, and information on the child consultation record system in advance). (For example, it may be an item extracted by analyzing past support data from 300. For example, it may be a checklist as shown in FIG. 6 or 7 of the risk assessment information.) Or, it may be an item entered by the user. information (for example, it may be entry information extracted from free descriptions filled in by the user, or it may be entry information extracted from entry items that are presented classified by predetermined support). )including.

The needs evaluation unit 135 performs a needs assessment evaluation regarding the support required by at least one of the child or a person related to the child, based on the registered child's basic information and reception data (especially needs assessment information). , generate evaluation data for each possible support). The generated evaluation data may be stored as reception data in association with the reception ID. The needs evaluation unit 135 receives, as input information, for example, past basic information of the child (age, gender, address (municipality)), reception information (type of abuse, reception hours, reception classification (new, re-notification, re-application), and Needs assessment information associated with reception information (reception route, main abuser, status of application to each item of needs assessment, etc.), information regarding whether or not limit assessment is applicable, image information about the child, information about the child's wounds, information about the child's wounds, etc. Psychological diagnosis information, information on the situation of the child's home, decision-making result information and/or safety confirmation result information, items related to corresponding implementation (whether temporary protection was implemented, whether temporary protection was canceled, whether the child was transferred to a child care facility, etc.) Whether measures have been implemented, whether measures have been lifted on child care facilities, etc., whether measures have been taken against foster parents, whether measures have been lifted against foster parents, whether cooperation with related organizations has been implemented, social resources and services. whether the local government or local community has established or used a monitoring system), information on mitate, information on future responses, and survey record information (survey date and time, survey title, agenda). Records (survey results (facts), additional information obtained from the investigation (risk assessment information (information obtained from the investigation), images of children obtained during the investigation, information on children's wounds obtained during the investigation, children obtained during the investigation) psychological diagnosis information and information on the child's home situation obtained during the survey, the date and time of the next interview for the survey, the status at the time of survey completion, the safety scale at the time of survey completion (an intuitive numerical value of the child's safety). When you input some or all information of input result information), response details, etc., the output information is similar to past cases (abuse is classified into multiple types, and the characteristics of the type that the current case is closest to are displayed. Needs assessment learning that outputs part or all of information such as the recommendation rate of each support based on the current case and similar cases (displaying past case records if there are past cases belonging to the type) The model may be used to generate evaluation data (output information).The learning model extracts at least a part of the above-mentioned input information and result information corresponding to the output information from past data to generate training data. The input information may be learned in advance, and may be stored in the storage unit 120, or may be stored in an external server. It is possible to output more precise output information by including the type of input information. As another example, the needs evaluation unit 135 extracts similar past cases according to a combination of some or all of the input information (particularly needs assessment information), and extracts the results of the extracted past cases. Some or all of the above output information may be generated by calculating statistical values from the above. Note that the methods for generating these various types of information and evaluation data are merely examples, and are not limited to these. Further, the needs evaluation unit 135 may generate evaluation data by further referring to at least one of survey record information and photo information, for example, as reception data to be referred to for evaluation.

Then, the needs evaluation unit 135 transmits the generated output information (evaluation data) to the user terminal 200, and displays the output information in a predetermined format on the user interface screen of the application on the user terminal 200. The text generation instruction unit 134 then uses the child identification information acquired from the storage unit 120 (for example, the reception data storage unit 122), the assessment information (assessment information of at least one of the risk assessment information and the needs assessment information), and the survey information. Text data (text generation instruction information) for instructing generation of a sentence based on at least recorded information (described later) is generated. The generated text data is stored in the storage unit 120 in association with at least child identification information (furthermore, in association with input date and time information, reception ID, etc.).

As described above, according to the present embodiment, by generating sentences using a trained language model, it is possible to reduce the burden on the person in charge at the site.

Although we have explained information regarding child abuse between child guidance centers and various local government agencies as an example, the information is not limited to this and can be any other type of information, but in particular information from public organizations such as municipalities and the national government. Information exchanged between parties (particularly information related to social issues such as domestic violence, sexual violence, bullying, harassment, crime, delinquency, physical disability, mental disability, developmental disability, child poverty, and elder abuse) The information may be information in areas related to food, clothing, shelter, and physical safety, or areas related to safety such as discrimination and abusive language, etc.), and the configuration may be such that corresponding information is managed depending on the target user. For example, (1) between public health centers or maternal and child health centers, between public health centers and maternal and child health centers, or between public health centers or maternal and child health centers and local governments, child information and family information collected at public health centers or maternal and child health centers, family childcare stress, etc. Investigation record information such as child care services used, etc. (2) Between domestic violence victim centers, domestic violence victim centers and local governments, etc., at least one of the perpetrators, victims, and bystanders obtained by domestic violence victim centers (3) Investigative record information regarding DV facts related to disability and welfare, obtained by personnel in charge between welfare centers or between welfare centers and local governments, etc., such as disability information and economic situation information, (4) Protection It may be used in relationships between probation offices, or between probation offices and local governments, such as investigation record information on crimes and trials obtained by probationers, and it can be used in relation to the relevant investigation items in each investigation. various indicators based on the occurrence (including reoccurrence) of domestic violence, sexual violence, bullying, harassment, crime, delinquency, child poverty, and elder abuse, or the mental and physical symptoms that may arise secondary to these. It may also be used to estimate the probability (probability of occurrence, etc.) related to the occurrence (including re-occurrence) of a failure.

The contents of the embodiments of the present invention will be listed and explained again. An information processing method, program, and information processing system according to an embodiment of the present invention has the following configuration.
[Item 1]
An information processing method for supporting child consultation work provided by a server connected to a user terminal of a user related to child consultation via a network,
The control unit of the server includes:
Accept at least as input information, from the user terminal, child identification information that identifies the child, assessment information for determining at least one of protection or support for the child, and investigation record information obtained through the investigation; storing the input information in a storage unit of the server;
A method for generating text data for instructing generation of a summary sentence based on at least the child identification information, the assessment information, and the investigation record information acquired from the storage unit.
[Item 2]
An information processing method for supporting child consultation work provided by a server connected to a user terminal of a user related to child consultation via a network,
The control unit of the server includes:
Accept at least as input information, from the user terminal, child identification information that identifies the child, assessment information for determining at least one of protection or support for the child, and investigation record information obtained through the investigation; storing the input information in a storage unit of the server;
A method of generating text data for instructing the generation of user support information estimated from the assessment information and the survey record information at one or more past points in time linked to the child identification information acquired from the storage unit. .
[Item 3]
A program that executes an information processing method to support child consultation work, which is provided by a server connected to the user terminal of a user related to child consultation via a network,
The control unit of the server includes:
Accept at least as input information, from the user terminal, child identification information that identifies the child, assessment information for determining at least one of protection or support for the child, and investigation record information obtained through the investigation; storing the input information in a storage unit of the server;
A program that generates text data for instructing generation of a summary sentence based on at least the child identification information, the assessment information, and the investigation record information acquired from the storage unit.
[Item 4]
A program that executes an information processing method to support child consultation work, which is provided by a server connected to the user terminal of a user related to child consultation via a network,
The control unit of the server includes:
Accept at least as input information, from the user terminal, child identification information that identifies the child, assessment information for determining at least one of protection or support for the child, and investigation record information obtained through the investigation; storing the input information in a storage unit of the server;
A program that generates text data for instructing generation of user support information estimated from the assessment information and the survey record information at one or more past points associated with the child identification information acquired from the storage unit. .
[Item 5]
An information processing system that supports child consultation services, including a server that connects to user terminals of users related to child consultation via a network,
The control unit of the server includes:
From the user terminal, at least child identification information that identifies the child, assessment information for determining whether to protect or support the child, and investigation record information obtained from the investigation are accepted as input information, and the storing the input information in a storage unit of the server;
An information processing system that generates text data for instructing generation of a summary sentence based on at least the child identification information, the assessment information, and the investigation record information acquired from the storage unit.
[Item 6]
An information processing system that supports child consultation services, including a server that connects to user terminals of users related to child consultation via a network,
The control unit of the server includes:
Accept at least as input information, from the user terminal, child identification information that identifies the child, assessment information for determining at least one of protection or support for the child, and investigation record information obtained through the investigation; storing the input information in a storage unit of the server;
Information for generating text data for instructing the generation of user support information estimated from the assessment information and the survey record information at one or more past points linked to the child identification information acquired from the storage unit. processing system.

Although the disclosed embodiments have been described above, they can be implemented in various other forms, and can be implemented with various omissions, substitutions, and changes. These embodiments and modifications, as well as omissions, substitutions, and changes, are included within the technical scope of the claims and their equivalents.

1 business support system, 100 first server, 110 communication unit, 120 storage unit, 130 control unit, 200 first user terminal, 300 child consultation record system, 400 text generation system, NW network

Claims (3)

An information processing method for supporting child consultation work provided by a server connected to a user terminal of a user related to child consultation via a network,
The control unit of the server includes:
Accept at least as input information, from the user terminal, child identification information that identifies the child, assessment information for determining at least one of protection or support for the child, and investigation record information obtained through the investigation; storing the input information in a storage unit of the server;
Generating a summary instruction text that includes the child identification information, the assessment information, and the investigation record information acquired from the storage unit as a summary target as prompt data that instructs generation of a summary text that summarizes the summary target;
presenting a summary sentence generated by the language model in response to the input of the prompt data to the user terminal;
Method. A program that executes an information processing method to support child consultation work, which is provided by a server connected to the user terminal of a user related to child consultation via a network,
The control unit of the server includes:
Accept at least as input information, from the user terminal, child identification information that identifies the child, assessment information for determining at least one of protection or support for the child, and investigation record information obtained through the investigation; storing the input information in a storage unit of the server;
Generating a summary instruction text that includes the child identification information, the assessment information, and the investigation record information acquired from the storage unit as a summary target as prompt data that instructs generation of a summary text that summarizes the summary target;
presenting a summary sentence generated by the language model in response to the input of the prompt data to the user terminal;
program. An information processing system that supports child consultation services, including a server that connects to user terminals of users related to child consultation via a network,
The control unit of the server includes:
Accept at least as input information, from the user terminal, child identification information that identifies the child, assessment information for determining at least one of protection or support for the child, and investigation record information obtained through the investigation; storing the input information in a storage unit of the server;
Generating a summary instruction text that includes the child identification information, the assessment information, and the investigation record information acquired from the storage unit as a summary target as prompt data that instructs generation of a summary text that summarizes the summary target;
presenting a summary sentence generated by the language model in response to the input of the prompt data to the user terminal;
Information processing system.