JP6955773B2 - Ability improvement means selection device, ability improvement means selection method, trained model, trained model generation program and ability improvement means output program - Google Patents

Description

The present invention relates to an ability-improving means selection device and a method of selecting an ability-improving means for selecting means for improving the ability of a person to be supported such as a disabled person, which is particularly suitable for employment support for persons with disabilities, and such ability-improving means. It relates to a trained model and a trained model generation program used for selection of, and an ability improvement means output program using the trained model.

In order for learners to improve their abilities, it is necessary to proceed with learning using appropriate teaching materials that match each person's abilities, but in order to respond to these issues, evaluation of the teaching materials used by learners An invention related to a system that collects information on learning results using teaching materials, analyzes this information with a computer, and presents teaching materials according to each learner's ability and a learning course that combines multiple teaching materials. Is disclosed (see, for example, Patent Documents 1 and 2).

Japanese Unexamined Patent Publication No. 2003-271752 JP-A-2017-134136

The invention relating to the system for selecting teaching materials, etc. as disclosed in Patent Documents 1 and 2 assumes that the learner is a healthy person, and it is natural that the learner can learn using the presented teaching materials. It is assumed that. However, when the target person becomes a person with a disability such as an intellectual disability, it is often not possible to proceed with learning simply by providing teaching materials.

For example, in the field of employment support for persons with disabilities such as employment transition support offices and special schools, for example, it is not possible to sit down and concentrate on the work, and it is not possible to proceed with the work in order just by being instructed by words. There are many people with disabilities who have such problems, and in order for these people with disabilities to acquire the abilities necessary for employment, it depends on the characteristics and situations of each individual so that they can work on teaching materials to improve their abilities. Reasonable accommodation such as individual adjustments and changes is required to eliminate the obstacles and difficulties that occur.

The present invention has been made in order to deal with such a problem, and is an ability improving means selection device for selecting a means for improving the ability of a person to be supported such as a person with a disability, which is particularly suitable for employment support for a person with a disability. An object of the present invention is to provide a method of selecting the ability-improving means, a trained model and a trained model generation program used for selecting such the ability-improving means, and an ability-improving means output program using the trained model. Is.

In order to solve such a problem, in the present invention, in the support such as employment support to be provided to improve the ability of the support target person, when the support target person learns using the teaching materials, each support target is provided. Reasonable accommodation will be given to enhance the learning effect according to the person, and the ability will be improved from the support implementation information including not only the teaching materials used by the support target person but also the reasonable accommodation given to the support target person. When generating a model or rule for selecting a means and formulating a support plan, by applying this model or rule to the support goal, not only the teaching materials according to each support goal but also the support target person The structure is such that the ability improvement means including reasonable accommodation can be selected to enhance the learning effect of.

The person to be supported in the present invention can be assumed to be a person with a disability who receives employment support, but the person to be supported is not limited to the person with a disability, and individual reasonable accommodation is given when learning the teaching materials. Can include a wide range of support recipients who need. In addition, for reasonable accommodation in the present invention, it is desirable for the support target person to give when learning the teaching material, and to remove the obstacles and difficulties that occur according to the characteristics and situations of each support target person. Consideration such as adjustment and change is applicable.

The invention relating to the ability improving means selection device for solving the problem according to the present application relates to the information about the support target person and the support regarding the support provided for improving the predetermined ability of the support target person who is a disabled person. Reasonable consideration given to enhance the learning effect of the learning of the teaching material of the support target person and the support when learning the teaching material for improving the ability of the support target person used in Machine learning is performed based on learning data including information on changes in the ability of the support target person before and after the implementation of, and learning of the teaching material suitable for each support goal and the teaching material of the support target person when learning the teaching material. on aid target comprising generating means for generating a learned model, information about the capacity desired to be improved by predetermined information and the support for supporting the subject to receive support for use in the selection of reasonable accommodation to enhance the learning effect by Information is input to the learned model generated by the generation means , and learning by learning the teaching material of the support target person when learning the teaching material suitable for the support target and the teaching material output from the learned model. It is a capacity improving means selection device comprising: a selection means for selecting a capacity improving means including rational consideration for enhancing the effect as a capacity improving means suitable for the support target.

What kind of reasonable accommodation is effective for each person to be supported varies according to the individuality of each person to be supported, especially when a person with a disability is the person to be supported, and rules should be established. Is not easy. Therefore, as described above, it is preferable to utilize the trained model generated by machine learning by using the information on the support provided in the past as the learning data.

Further, the generation means may be characterized in generating the trained model using information about a predetermined ability goal to be improved by the support in the support plan formulated before the support is implemented. ..

In this way, if the generation of the model or rule reflects not only the change in the ability of the support recipient before and after the support implementation but also the goal of the ability in the support plan, more detailed information analysis can be performed. , It is an effective configuration to improve the accuracy of selecting the ability improvement means suitable for each support.

Further, in the selection means, the information about the support target input to the trained model and the information about the support plan including the ability improvement means output from the trained model and selected in the information about the support target are stored. When the information regarding the change in the predetermined ability of the support target person after the support based on the support plan is received, the generation means reflects the information regarding the support plan and the information regarding the change in the ability. It may be characterized by updating the trained model.

With this configuration, the information on the support performance regarding the support plan using the capacity improvement means selected by the capacity improvement means selection device according to the present invention is further reflected in the model or the rule to promote the evolution. Alternatively, the configuration is suitable for improving the accuracy of the rule.

The present invention can also be specified as an invention of a method for selecting a capacity improving means, which is executed by the capacity improving means selecting device according to the present invention.

In the method of selecting the ability improving means according to the present invention, the ability improving means selection device for selecting the ability improving means for improving the predetermined ability of the support target person who is a disabled person improves the predetermined ability of the support target person. Regarding the support provided for the support, the information about the support target person, the teaching material for improving the ability of the support target person used for the support, and the teaching material of the support target person when learning the teaching material. Machine learning is performed based on the rational consideration given to enhance the learning effect of the learning and the learning data including the information on the change of the ability of the support target person before and after the support is provided, and each support is performed. A generation step for generating a learned model used for selecting a teaching material suitable for a goal and a rational consideration for enhancing the learning effect of the learning of the teaching material of a support target person in learning the teaching material, and the ability improving means selection device. but the information about the assistance target including information on capacity desired to be improved by predetermined information and the support for supporting the subject for assistance, and input to the trained model generated in said generating step, from the learned model The output teaching material suitable for the support target and the ability improving means including rational consideration for enhancing the learning effect of the learning of the teaching material of the support target person when learning the teaching material are the ability suitable for the support target. It is a method of selecting an ability improving means characterized by having a selection step to be selected as an improving means.

Further, the present invention can also be specified as an invention of a method for selecting a capacity improving means, which is executed by each configuration of the capacity improving means selecting device according to the present invention described above.

The present invention can also be specified as an invention of a learned model used in the ability improving means selection device according to the present invention.

In the learned model according to the present invention, regarding the support provided for improving the predetermined ability of the support target person who is a disabled person, the information about the support target person and the support target person used for the support and reasonable accommodation made to enhance the learning effect of the materials of learning of the assistance subject upon the materials and the materials for improving the capability learning, the support subject before and after carrying out said support Information about the support target, which is generated by performing machine learning based on the learning data including the information about the change of the ability, and includes the predetermined information about the support target person who receives the support and the information about the ability to be improved by the support. When input, in order to make the computer function so as to output the teaching material suitable for the support target and the ability improving means including the rational consideration for enhancing the learning effect of the learning of the teaching material of the support target person when learning the teaching material. Is a trained model of.

Further, the present invention can also be specified as an invention of a learned model used in each configuration of the ability improving means selection device according to the present invention described above.

The present invention can also be specified as an invention of a program for generating a trained model used in the ability improving means selection device according to the present invention.

The learned model generation program according to the present invention was used for the support provided to the computer in order to improve the predetermined ability of the support target person, the information about the support target person, and the support. Reasonable consideration given to improve the learning effect of the teaching material for improving the ability of the support target person and the learning effect of the teaching material for the support target person when learning the teaching material, and before and after the support is provided. A support goal including predetermined information about a support target person who receives support by executing machine learning based on learning data including information about a change in the ability of the support target person in the above and information about an ability to be improved by the support. When the information about the teaching material is input, the learning model that outputs the teaching material suitable for the support goal and the ability improving means including the rational consideration for enhancing the learning effect of the learning of the teaching material of the support target person when learning the teaching material is generated. It is a trained model generation program characterized by generating.

Further, the present invention can also be specified as an invention of a program for generating a trained model used in each configuration of the ability improving means selection device according to the present invention described above.

The present invention can also be specified as an invention of a program that outputs the ability-improving means using a learned model, which is executed in the ability-improving means selection device according to the present invention.

The ability-improving means output program according to the present invention was used for the support provided to the computer in order to improve the predetermined ability of the support target person, the information about the support target person, and the support. Reasonable consideration given to improve the learning effect of the teaching material for improving the ability of the support target person and the learning effect of the teaching material for the support target person when learning the teaching material, and before and after the support is provided. We would like to improve the trained model generated by performing machine learning based on the learning data including the information on the change of the ability of the support target person in the above by the predetermined information on the support target person to be supported and the support. Ability improvement including rational consideration for accepting input of information on support goals including information on abilities and enhancing the learning effect of the learning of the teaching materials of the support target person when learning the teaching materials suitable for the support goals and the teaching materials. It is a capacity improving means output program characterized by executing a process of outputting means.

Further, the present invention can also be specified as an invention of a program that outputs the ability improving means using the learned model, which is executed in each configuration of the ability improving means selection device according to the present invention described above.

According to the present invention, not only a teaching material suitable for improving the ability of each support target person, but also a reasonable accommodation according to the characteristics of each support target person is selected as a means for improving the ability of the support target person. It will be possible to select more effective ability-improving means for persons with disabilities, etc., for whom reasonable accommodation is desirable as a premise of learning by.

Since the present invention is particularly suitable for employment support for persons with disabilities and can be expected to promote the improvement of abilities required for employment of persons with disabilities, it is considered that the present invention has great social significance.

It is a figure which shows the outline of embodiment of the capacity improvement means selection apparatus which concerns on this invention. It is a figure which shows the functional structure of the capacity improvement means selection apparatus in 1st Embodiment of this invention. It is a figure which shows the functional structure of the capacity improvement means selection apparatus in 2nd Embodiment of this invention. It is a figure which shows an example of the evaluation information of the support target person stored as a part of the target person information in the ability improvement means selection apparatus which concerns on this invention. It is a figure which shows the flow of making and carrying out the support plan using the capacity improvement means selection apparatus which concerns on this invention. It is a figure which shows the 1st example of the support implementation information stored in the capacity improvement means selection apparatus which concerns on this invention. It is a figure which shows the 2nd example of the support implementation information stored in the ability improvement means selection apparatus which concerns on this invention. It is a figure which shows the 1st example of selection of the capacity improvement means by the capacity improvement means selection apparatus which concerns on this invention. It is a figure which shows the 2nd example of selection of the capacity improvement means by the capacity improvement means selection apparatus which concerns on this invention.

The embodiment for carrying out the present invention will be described in detail below with reference to the drawings. In the following description, an example will be described in which the present invention is applied to employment support for persons with disabilities and a learned model generated by machine learning is used for selecting a means for improving ability. This is an example of the embodiment, and the present invention is not limited to the embodiments described below.

FIG. 1 shows an outline of an embodiment of a capacity improving means selection device according to the present invention. The ability-improving means selection device according to the present invention was implemented using teaching materials while giving reasonable consideration to the support target persons such as persons with disabilities who receive employment support in order to enhance the learning effect of the support target persons. Support implementation information related to support is accumulated, and the support implementation information is used as learning data in a trained model generated by machine learning (a rule base may be adopted instead of machine learning) to support the support recipients. When information about the goal is input, teaching materials suitable for the support goal and ability improvement means including rational consideration for enhancing the learning effect of the support target person are output.

When selecting teaching materials for healthy people, it is a natural premise to study using the presented teaching materials, so it is sufficient to output information on teaching materials suitable for the goal, but support targets When a person becomes a person with a disability such as an intellectual disability, it is often not possible to proceed with learning simply by providing teaching materials. Therefore, in the present invention, in addition to the teaching material, reasonable accommodation is added to the information output as a means for improving the ability to enhance the learning effect of the support target person, and also as the learning data that is the basis for generating the learned model. It is characterized by adding information not only about which teaching material was used, but also what kind of reasonable accommodation was given.

In the present invention, which is supposed to be used for employment support for persons with disabilities, the teaching materials used for improving the ability of the support recipient include general teaching materials such as textbooks and drills, as well as stamps and a set of materials. Routine work such as, living in a group home, and programs such as one-on-one social skill training will also be included.

Further, the reasonable accommodation for enhancing the learning effect of the support target person in the present invention includes individual adjustments and changes for removing obstacles and difficulties that occur according to the characteristics and situations of the support target person. With reasonable accommodation, for example, provide an individual whiteboard that can be kept and used by the support recipient to describe the work process, check the completed process, or support target. Consideration using tools (individual whiteboards, tapes, jigs that can check numbers, etc.) such as attaching tape to the floor so that people can see where they are sitting and providing jigs that can check numbers, and siblings It includes environmental considerations such as living in a remote environment.

The capacity improving means selection device according to the present invention having the above characteristics is implemented as a computer operated standalone (first embodiment) or as a server connected to a network (second embodiment). Form) is also possible.

In the first embodiment, from an input / output device such as a computer keyboard or monitor installed at a site of employment support for persons with disabilities such as a employment transition support office or a special needs school, which functions as a capacity improvement means selection device. When the support implementation information is input and stored in the computer, and the information on the support target is input from the input / output device, the capacity improving means suitable for the support target is selected and output to the input / output device.

In the second embodiment, a PC that is connected to a server that functions as a capacity improving means selection device via a network such as the Internet and is installed at a work support site for persons with disabilities such as a work transition support office or a special needs school. When the support implementation information is input from the business terminal such as (personal computer) and sent to the server, the support implementation information is accumulated in the server. Further, when the information on the support target is input to the business office terminal and transmitted to the server, the information on the capacity improving means suitable for the support target output from the server is received by the business office terminal.

FIG. 2 shows the functional configuration of the capacity improving means selection device in the first embodiment described above, and the functional configuration of the capacity improving means selection device in the second embodiment. Is shown in FIG.

In FIG. 2, which shows the configuration in the first embodiment, the capacity improving means selection device 10 corresponds to the capacity improving means selection device according to the present invention. The computer functioning as the capacity improving means selection device 10 is provided with an auxiliary storage device such as a CPU, a main memory, and an HDD, and a program stored in the auxiliary storage device is read into the main memory and the CPU executes arithmetic processing. By doing so, a predetermined function is realized.

The physical configuration of the computer that functions as the capacity improving means selection device 10 is not particularly limited, and in addition to the function of selecting the capacity improving means in the present invention, other functions are provided in the same computer. There may be. Further, each function required for the present invention may be physically realized by one computer, or may be realized by using a plurality of computers.

The support implementation information reception unit 11, the machine learning execution unit 14, the support target reception unit 16, the capacity improvement means selection unit 17, and the capacity improvement means output unit 18 of the capacity improvement means selection device 10 are all functionally specified. A program corresponding to the function of each part stored in an auxiliary storage device such as an HDD is read into the main memory, and the CPU executes arithmetic processing to realize the function corresponding to each part.

The target person information storage unit 12, the support implementation information storage unit 13, and the learned model storage unit 15 of the capacity improvement means selection device 10 include auxiliary storage devices such as HDDs provided in the computer constituting the capacity improvement means selection device 10. Storage space is allocated. These storage areas are not required to be physically provided in one computer, and may be distributed in a plurality of computers.

An input / output device to a computer such as a keyboard or a monitor is used as the input / output device 19, but if the device can input / output data to / from the computer constituting the capacity improving means selection device 10, the configuration is It is not particularly limited.

In FIG. 3, which shows the configuration in the second embodiment, the capacity improving means selection server 20 corresponds to the capacity improving means selection device according to the present invention. The computer functioning as the capacity improving means selection server 20 is provided with an auxiliary storage device such as a CPU, a main memory, and an HDD, and a program stored in the auxiliary storage device is read into the main memory and the CPU executes arithmetic processing. By doing so, a predetermined function is realized.

The physical configuration of the computer functioning as the capacity improving means selection server 20 is not particularly limited as long as it can be connected to a network such as the Internet. In addition, other functions may be provided in the same computer. Each function required for the present invention may be physically realized by one computer, or may be realized by using a plurality of computers.

The support execution information reception unit 21, the machine learning execution unit 24, the support target reception unit 26, the capacity improvement means selection unit 27, and the capacity improvement means transmission unit 28 of the capacity improvement means selection server 20 are all functionally specified. A program corresponding to the function of each part stored in an auxiliary storage device such as an HDD is read into the main memory, and the CPU executes arithmetic processing to realize the function corresponding to each part.

The target person information storage unit 22, the support implementation information storage unit 23, and the learned model storage unit 25 of the capacity improvement means selection server 20 include auxiliary storage devices such as HDDs provided in the computer constituting the capacity improvement means selection server 20. Storage space is allocated. These storage areas are not required to be physically provided in one computer, and may be distributed in a plurality of computers.

A network terminal that can be connected to a network such as the Internet, such as a PC or a tablet computer, is used as the office terminal 30, but if it is a terminal that can perform data communication with the capacity improving means selection server 20, the configuration thereof. Is not particularly limited.

On the premise of the above configuration, the method of selecting the capacity improving means by the capacity improving means selection device according to the present invention will be specifically described with reference to FIG. 4-9. In the following, the description will be made on the premise of the first embodiment whose configuration is shown in FIG. 2, but even in the second embodiment whose configuration is shown in FIG. 3, the input / output of information is the input / output device 19 Instead, it is only performed at the business office terminal 30 via a network such as the Internet, and the processing executed by the capacity improving means selection server 20 is substantially the same as the processing executed by the capacity improving means selection device 10. Is the same.

In the present invention, information about a person with a disability who receives employment support as a support target person is input from the input / output device 19 by a supporter or the like in charge of employment support, and the target person information is stored via the support implementation information reception unit 11. It is stored in the unit 12. The content of the information about the disabled person stored in the target person information storage unit 12 is not particularly limited, but it is a basic matter such as the name, gender, age, address, family structure and support system of the disabled person. In addition to personal data, for example, it can include evaluation information that evaluates disability characteristics based on examinations and diagnoses, lifestyles and working habits of persons with disabilities, PC skills, and the like.

FIG. 4 shows an example of a data structure regarding evaluation information of a disabled person who is a support target person, which is stored in the target person information storage unit 12. In the target person information storage unit 12, evaluation items such as basic evaluations related to the living and personality aspects of persons with disabilities, employment evaluations related to work habits and work performance, and professional evaluations related to PC skills are associated with the tree structure. , Scores evaluated for each item are stored.

The evaluation of persons with disabilities who are the target of support is performed regularly, for example, once a quarter, and the evaluation information for each person with disabilities who is the target of support is stored in the target person information storage unit 12. It is stored. The evaluation information stored in the target person information storage unit 12 corresponds to the evaluation described in natural sentences and the evaluation expressed in standard sentences, in addition to the evaluation indicating the level in the predetermined evaluation item by the score. It can include an evaluation indicating whether or not to do so. The score of the evaluation item at the top of the tree may be automatically calculated based on, for example, the average value of the scores of the evaluation items belonging to the bottom of the evaluation item, the ratio of the number of evaluation items exceeding a predetermined threshold value, and the like. good.

FIG. 5 shows the flow of creating and implementing a support plan using the capacity improving means selection device 10. First, the flow of accumulating support implementation information and generating a learned model shown in S1-S4. Will be described.

Persons in charge of employment support for persons with disabilities at employment transition support offices, etc., at the timing of regular evaluation such as after the support according to the support plan is completed or once a quarter (the timing of input is particularly limited). (Not), the support implementation information regarding the results of the support provided to the disabled person who is the support target person is input from the input / output device 19 (S1). The input information is received by the support implementation information reception unit 11, and is stored in the support implementation information storage unit 13 together with the information read from the target person information storage unit 12. By repeating such an operation for a plurality of the provided support, the support implementation information is accumulated in the support implementation information storage unit 13 (S2).

6 and 7 show an example of the support implementation information stored in the support implementation information storage unit 13. From the input / output device 19, the evaluation items and the support for which the ability is improved by the support are shown. Enter information on the identification information of the support recipient, the teaching materials used in the support, and the reasonable accommodation made to enhance the learning effect of the support target in the support at the two points of time to identify before and after the support. Will be done.

Of these, for the evaluation items whose abilities were improved by support, one of the evaluation items stored as evaluation information in the target person information storage unit 12 (in the example shown in FIG. 4, the evaluation is based on the score or natural text. (Evaluation item in which is recorded) is selected, the identification information of the support target person and the change in ability related to the evaluation item before and after the support is provided from the two points of time to identify before and after the support. Can be identified.

In the example of Fig. 6, is it possible to take a seat with respect to the concentration required for work preparation (in the example of Fig. 4, "prospect" included in the evaluation items of "professional evaluation"> "work preparation"> "concentration" Corresponding to the evaluation that "I can't take a seat if I don't understand") and how much I can understand the process for work execution (in the example of Fig. 4, it is included in the evaluation items of "work evaluation"> "work execution"> "process understanding" (Corresponding to the evaluation that "word instructions do not lead to process understanding") is specified as an evaluation item, and for Mr. A specified from the identification information of the support target person, before the support specified from time point 1, each " The evaluations of "I can't take a seat without knowing the outlook" and "The instructions of words do not lead to understanding the process" have changed to the evaluations of "I can take a seat" and "I can understand the process" after the support identified from point 2. Can be specified from the evaluation information of the support target person A stored in the target person information storage unit 12. If the timing of inputting the above-mentioned information overlaps with the input of the post-support evaluation information specified from the time point 2, the post-support evaluation information will be input at this timing.

Further, from the input identification information of the support target person, it is possible to specify the personal data, failure characteristics, evaluation information, etc. of the support target person A stored in the target person information storage unit 12. Among them, the information about the support target person used for generating the trained model to be described later is selected and read from the target person information storage unit 12.

As described above, the change in the ability of the support target person before and after the support is provided, which is specified by the evaluation items input from the input / output device 19 and the information (time point 1, time point 2) that identifies the time points before and after the support. Information about the support target person specified from the support target person's identification information input from the input / output device 19, the teaching materials used for the support input from the input / output device 19, and the learning effect of the support target person are enhanced. The support implementation information consisting of rational consideration made for this purpose is stored in the support implementation information storage unit 13.

The items of information included in the support implementation information stored in the support implementation information storage unit 13 are not limited to the example shown in FIG. Information on support goals such as evaluation target values specified in the support plan may be added.

Further, all the support implementation information stored in the support implementation information storage unit 13 is directly read from the input / output device 19 without reading some information included in the support implementation information from the target person information storage unit 12. It may be configured to be input.

In the example of FIG. 7, in order to enhance the learning effect of the information on the change in the ability of the support target person before and after the support, the information on the support target person, the teaching material used for the support, and the support target person by the same processing. The support implementation information consisting of reasonable accommodations made is stored in the support implementation information storage unit 13, but in this example, the evaluation item (“process understanding”” included in the evaluation information stored in the target person information storage unit 12. And "Understanding the numbers") will show the change in the ability of the support recipient before and after the support is provided.

As described above, support implementation information showing what kind of teaching materials and what kind of rational consideration was given to the support recipients with what characteristics, and what abilities changed and how. Is stored in the support implementation information storage unit 13, and the machine learning execution unit 14 performs machine learning using this information as learning data (S3), and each support target (what evaluation item's ability is what degree). A learned model used for selecting teaching materials suitable for (the goal of improving) and rational consideration for enhancing the learning effect of the support recipient is generated (S4) and stored in the learned model storage unit 15. ..

Of the support implementation information, for information that overlaps with other support implementation information such as information about the support target person, the support implementation information stored in the support implementation information storage unit 13 includes identification information of the support target person, etc. Only the data used as the key of the above may be recorded, and the necessary information may be read out from the target person information storage unit 12 when the machine learning execution unit 14 executes the machine learning.

The trained model generated in S4 executes an operation based on the weighting coefficient of the neural network for the support target input to the input layer of the neural network, and outputs information on the capability improving means from the output layer of the neural network. It is a trained model that makes the computer function so that the weighting coefficient applied here is information on changes in ability before and after support implementation, information on support recipients, teaching materials used for support, and rational consideration. It is obtained by machine learning using the support implementation information including it as learning data.

As explained above, other information may be added to the information contained in the support implementation information used as learning data, but for example, it is defined in the support plan formulated before the support is implemented. When information about support goals such as evaluation target values is added, information about the level at which the goals are set and learned with teaching materials is added to the learning target of the learned model, so more detailed trends and correlations are added. It can be expected that the relationship will be reflected in the model and the accuracy of selecting the ability-improving means suitable for each support will be improved.

Next, a flow of inputting a support goal into the trained model generated in S4 and outputting information on the ability improving means described in S5-S7 will be described.

When creating a support plan for persons with disabilities who are the target of support, the person in charge of employment support for persons with disabilities at the employment transition support office, etc. inputs information on the support target for the support from the input / output device 19 (S5). ). When the input information about the support goal is received by the support goal reception unit 16, the ability improvement means selection unit 17 is activated, and the information about the support goal is input to the learned model stored in the learned model storage unit 15. (S6). From the trained model, teaching materials suitable for the support goal and ability improvement means including reasonable accommodation for enhancing the learning effect of the support target person are output, and these teaching materials and reasonable accommodation are the ability improvement means selection device 10. As information about the capacity improving means selected by, the capacity improving means output unit 18 outputs the information to the input / output device 19 (S7).

8 and 9 show an example in which information on the support target is input to the trained model stored in the trained model storage unit 15 and information on the ability improving means is output. From the device 19, an evaluation item for which the ability of the support target person is to be improved by support, a target such as a score of the evaluation item to be improved by support, and identification information of the support target person are input.

Of these, for the evaluation items for which the ability of the support target person is to be improved by support, one of the evaluation items stored as evaluation information in the target person information storage unit 12 (in the example shown in FIG. 4, the score and nature). If the evaluation item (evaluation item in which the evaluation is recorded in the sentence) is selected, it is possible to specify the ability of the support target person regarding the evaluation item at the latest time from the identification information of the support target person, but separately. The difference from the target such as the score of the evaluation item that you want to improve by the input support shows what kind of ability you want to improve by the support and how much.

In the example of Fig. 8, is it possible to take a seat with respect to the concentration required for work preparation (in the example of Fig. 4, "prospect" included in the evaluation items of "professional evaluation"> "work preparation"> "concentration" Corresponding to the evaluation that "I can't take a seat if I don't understand") and how much I can understand the process for work execution (in the example of Fig. 4, it is included in the evaluation items of "work evaluation"> "work execution"> "process understanding" (Corresponding to the evaluation that "word instructions do not lead to process understanding") is specified as an evaluation item, and each evaluation is "you cannot sit down unless you understand the outlook" and "word instructions do not lead to process understanding". However, it is possible to specify from the latest evaluation information of Mr. C, which is stored in the target person information storage unit 12 and is specified from the identification information of the support target person. On the other hand, as the post-support goals of the evaluation item that we want to improve by support, we accept the input of evaluations such as "take a seat" and "understand the process", respectively.

Further, from the input identification information of the support target person, it is possible to specify the personal data, failure characteristics, evaluation information, etc. of the support target person C stored in the target person information storage unit 12. Among them, the information about the support target person who needs to be input to the trained model is selected and read from the target person information storage unit 12.

The trained model storage unit 15 is provided with predetermined information about the support target person and information about the support target including information about the ability to be improved by the support (information about the current evaluation and the target evaluation regarding the predetermined evaluation item). When input to the memorized trained model, the trained model outputs teaching materials suitable for the support goal and ability-improving means including rational consideration for enhancing the learning effect of the support target person.

In the example of FIG. 8, Mr. C's current evaluations of "I can't take a seat without knowing the outlook" and "Word instructions do not lead to understanding the process" are improved to "I can take a seat" and "I can understand the process", respectively. In response to the support goal of letting Mr. C take into consideration the characteristics of Mr. C, he learned the teaching material "seal stamping", and when learning, "tape the floor where he sits and returns" and "individually". A reasonable accommodation, such as providing a whiteboard, writing a time schedule, and checking when finished, is output as a means of improving capacity that can be expected to be effective against support goals.

In the example of FIG. 9, Mr. D's current evaluation that the score of "working habit" is "2" and the score of "mental aspect" is "2" is evaluated, and the scores of "working habit" are "3" and "mental", respectively. For the support goal of improving the score of "face" to "3", after considering Mr. D's characteristics, let him learn the teaching material "group home life", and in that learning, "supporter" Reasonable consideration such as "fixing" and "living in an environment away from the older brother" is output as a means of improving ability that can be expected to be effective against the support goal.

In the explanation so far, an example in which a trained model is generated in S1-S4 of FIG. 5 and information about a support goal is input to the trained model in S6 has been described, but a rule base is used instead of the trained model. In S1-S4, the support implementation information is used to generate a rule for selecting the ability improvement means suitable for the support goal, and in S6, the rule is applied to the information on the support goal, and the ability suitable for the support goal is applied. The improvement means may be output.

At the support site, a support plan for the support target person is created from the support target input to the capacity improvement means selection device 10 and the capacity improvement means selected and output by the capacity improvement means selection device 10 by inputting the support target and the like. Then (S8), support based on the support plan will be implemented (S9). After the support is implemented, the result is input to the capacity improvement means selection device 10 as new support implementation information (S1), but it corresponds to the capacity improvement means output from the capacity improvement means output unit 18. If the information about the support target and the support target person is also stored in the support implementation information storage unit 13, only the evaluation after the support needs to be input when the support implementation information is input, so that the support implementation information is accumulated. The processing to be done is streamlined. In this way, if the information on the support performance corresponding to the support plan created by using the capacity improvement means output from the trained model is further reflected in the trained model and its evolution is promoted, it is assumed. The configuration is also suitable for improving the accuracy of the trained model.

10 Ability improvement means selection device 11 Support implementation information reception unit 12 Target person information storage unit 13 Support implementation information storage unit 14 Machine learning execution unit 15 Learned model storage unit 16 Support target reception unit 17 Ability improvement means selection unit 18 Ability improvement Means output unit 19 Input / output device 20 Capacity improvement means selection server 21 Support implementation information reception unit 22 Target person information storage unit 23 Support implementation information storage unit 24 Machine learning execution unit 25 Learned model storage unit 26 Support target reception unit 27 Capacity Improvement means selection unit 28 Capacity improvement means transmission unit 30 Business office terminal

Claims (7)

Regarding the support provided to improve the predetermined ability of the support target person who is a disabled person, the information about the support target person, the teaching material for improving the ability of the support target person used for the support, and the above. Includes rational consideration given to the support target person in learning the teaching material to enhance the learning effect of learning the teaching material, and information on the change in the ability of the support target person before and after the support is provided. A trained model that performs machine learning based on learning data and is used to select teaching materials suitable for each support goal and rational consideration to enhance the learning effect of the learning of the teaching materials of the support target person when learning the teaching materials. And the generation means to generate
Information about support targets including information on capacity desired to be improved by predetermined information and the support for supporting the subject for assistance, type the learned model the generation unit has generated, is output from the learned model the ability improving means including reasonable accommodation to enhance the learning effect of the materials of learning of the assistance subject upon learning of materials and the materials suitable for the support target, the ability improved means suitable for the support target The selection method to select and
A capacity-enhancing means selection device characterized by comprising. Said generating means, according to claim, characterized in that by using the information related to the target of a given capacity to be improved by the support in which was formulated support plan prior to performing said support, it generates the learned model 1 The capacity-improving means selection device described. In the selection means, the information about the support goal input to the trained model and the information about the support plan including the ability improvement means output from the trained model and selected in the information about the support goal are stored.
When the information regarding the change in the predetermined ability of the support target person after the support based on the support plan is received, the generation means reflects the information regarding the support plan and the information regarding the change in the ability, and the above-mentioned The ability-enhancing means selection device according to claim 1 or 2, wherein the trained model is updated. The support target is the support provided by the ability improvement means selection device for selecting the ability improvement means for improving the predetermined ability of the support target person who is a disabled person in order to improve the predetermined ability of the support target person. It was carried out in order to enhance the learning effect of the learning of the teaching material of the support target person when learning the information about the person, the teaching material for improving the ability of the support target person used for the support, and the teaching material. Machine learning is performed based on learning data including rational consideration and information on changes in the abilities of the support target person before and after the support is provided, and support is provided when learning the teaching materials suitable for each support goal and the teaching materials. A generation step to generate a trained model used to select rational consideration to enhance the learning effect of the subject 's learning of the teaching material, and a generation step.
The ability improving means selection device, information about the assistance target including information on capacity desired to be improved by predetermined information and the support for supporting the subject for assistance, and input to the trained model generated in said generating step the output from the learned model, the ability improvement means including a reasonable accommodation to enhance the learning effect of the materials of learning of the assistance subject upon learning of materials and the materials suitable for the support target, the Selection steps to select as a capacity-building tool suitable for support goals,
A method of selecting a capacity-improving means characterized by having. Regarding the support provided to improve the predetermined ability of the support target person who is a disabled person, the information about the support target person, the teaching material for improving the ability of the support target person used for the support, and the above. Includes reasonable accommodations made to enhance the learning effect of the support target person by learning the teaching material when learning the teaching material, and information on the change in the ability of the support target person before and after the support is provided. Generated by performing machine learning based on training data
When the information about the support goal including the predetermined information about the support target person who receives the support and the information about the ability to be improved by the support is input, the teaching material suitable for the support goal and the teaching material of the support target person when learning the teaching material. A trained model for making a computer function so as to output ability-improving means including rational consideration for enhancing the learning effect of learning. Regarding the support provided to the computer to improve the predetermined ability of the support target person who is a disabled person, the information about the support target person and the ability of the support target person used for the support are to be improved. Regarding the rational consideration given to enhance the learning effect of the learning of the teaching material of the support target person when learning the teaching material and the teaching material, and the change of the ability of the support target person before and after the support is provided. When machine learning is executed based on learning data including information and information on a support target including predetermined information on a support target person receiving support and information on the ability to be improved by the support is input, it is suitable for the support target. A trained model generation characterized by generating a trained model that outputs an ability-improving means including rational consideration for enhancing the learning effect of the learning of the teaching material of the support target person when learning the teaching material and the teaching material. program. Regarding the support provided to the computer to improve the predetermined ability of the support target person who is a disabled person, the information about the support target person and the ability of the support target person used for the support are to be improved. Regarding the rational consideration given to enhance the learning effect of the learning of the teaching material of the support target person when learning the teaching material and the teaching material, and the change of the ability of the support target person before and after the support is provided. In the trained model generated by performing machine learning based on learning data including information, input information about a support target including predetermined information about a support target person receiving support and information about an ability to be improved by the support. Is received, and when learning the teaching material suitable for the support target, the process of outputting the ability improving means including the rational consideration for enhancing the learning effect of the learning of the teaching material of the support target person is executed. Ability improvement means output program.

Images