JP6963454B2 - Learning continuation system and learning continuation method - Google Patents

Description

The present invention relates to a learning continuation system and a learning continuation method.

In recent years, adaptive learning has been used to provide educational services according to the learning situation of each learner. Adaptive learning supports learners in achieving their goals by analyzing the learning history of each learner in real time and automatically providing individually optimized problems and explanations (see Non-Patent Document 1).

Technical commentator, "[Z-kai Career Advancement Course] TOEIC (R) Test Adapter [All levels] Starts on March 28! ◆ 20% OFF! Until March 30 during the start commemorative campaign", 2016 March 29, 2016, [online], [Search March 29, 2016], Internet <URL: http://www.news2u.net/releases/144042>

However, although adaptive learning can optimize the learning efficiency of learners, it has been difficult to achieve objectives other than optimizing learning efficiency. In addition to adaptive learning, it is left to the discretion of the instructor and other individuals to take appropriate measures that can encourage the continuation of educational services that are the basis of learning support for learners, in addition to optimizing learning efficiency. It wasn't planned.

The present invention has been made in view of this viewpoint, and an object of the present invention is to provide a learning continuation system and a learning continuation method that provide appropriate measures that can promote the continuation of educational services for each target person such as a learner. It is a thing.

In order to solve the above-mentioned problems, the learning continuation system from the first viewpoint is
A memory that stores behavioral characteristics and learning history for each of a plurality of subjects, as well as withdrawal or ongoing situations , and measures taken for each subject.
To build a predictive model leading to withdrawal of the subject of the withdrawal already based on the behavioral characteristics and the history of learning before Symbol subject of the withdrawal already, withdrawal index for behavioral characteristics and learning history on the basis of the predictive model Withdrawal from the time of calculation of the past withdrawal index for the ongoing subject whose withdrawal index is larger than the threshold value and the subsequent withdrawal index is equal to or less than the threshold value. It is characterized by having a control unit that reads out the measures taken up to the time of calculating the index from the memory , evaluates the measures, and presents the measures to be taken based on the evaluation result.

In addition, in the learning continuation system from the second viewpoint,
The learning history includes the status of the subject's efforts for the causatives provided to the subject.
It is preferable that the control unit presents the action status as a measure for improving the causative based on the action status when the correlation with the withdrawn status is detected.

In addition, in the learning continuation system from the third viewpoint,
Construction of the prediction model, the withdrawal already situations, as well as the detection of the correlation of the behavioral characteristics and the history of learning before Symbol subject's dehydration Shisasumi, said detected withdrawal already said on the basis of the correlation It is preferable to include the behavioral characteristics leading to the withdrawal of the subject and the estimation of the learning history.

In addition, in the learning continuation system from the fourth viewpoint,
The estimation of the behavior characteristics and the learning history leading to withdrawal of withdrawal already subjects, the plurality of subjects for each, based on the behavioral characteristics and the learning history is detected correlation with the state of the withdrawal already preferably includes a calculation of the withdrawal index.

In addition, in the learning continuation system from the fifth viewpoint ,
Prior Symbol controller, in the evaluation of the measures run against the subject of the ongoing based on the prediction model, the withdrawal exponent is stored in the memory, past the withdrawal index than the latest read out from said memory Is larger than the threshold value and the latest withdrawal index is equal to or less than the threshold value. For each of the behavioral characteristics and the learning history in which the correlation with the read-out and withdrawal status is detected, the number of the target persons who have performed the implemented measure is relative to the counted number of target persons. It is preferable to calculate the evaluation index evaluated based on the ratio.

In addition, in the learning continuation system from the sixth viewpoint,
In presenting measures to be taken based on the evaluation result, the control unit gives the target person whose latest withdrawal index is larger than the threshold value the evaluation index for the behavioral characteristics and the learning history of the target person. It is preferable to present the higher-ranked measures taken as the measures to be taken.

In addition, in the learning continuation system from the seventh viewpoint,
The memory stores the behavioral characteristics and the learning history, and the withdrawn or ongoing situation at each time period.
It is preferable that the control unit detects the correlation at each time and presents the measures to be taken based on the behavioral characteristics and learning history in which the correlation is detected at each time.

As described above, the means for solving the present invention have been described as an apparatus, but the present invention can also be realized as a storage medium in which a method, a program, and a program substantially corresponding thereto have been recorded, and the scope of the present invention. It should be understood that these are also included in.

For example, a learning continuation method that realizes the seventh aspect of the present invention as a method is
It is a learning continuation method that is executed by a computer.
A memory step that memorizes the behavioral characteristics and learning history of each of a plurality of subjects, the status of withdrawal or ongoing , and the measures taken for each subject.
The construction steps based on the behavioral characteristics and the history of learning before Symbol subject of the withdrawal already, to build a prediction model leading to withdrawal of the subject of the withdrawal already,
Based on the prediction model, a calculation step for calculating the withdrawal index for behavioral characteristics and learning history at different times, and
From the time when the past withdrawal index is calculated to the time when the subsequent withdrawal index is calculated for the ongoing subject whose withdrawal index calculated in the past is larger than the threshold value and the subsequent withdrawal index is equal to or less than the threshold value. The evaluation step to read the measures taken in the memory from the memory and evaluate the measures,
It is characterized by having a presentation step that presents the measures to be taken based on the evaluation result.

According to the present invention, it is possible to provide an appropriate measure that can promote the continuation of educational services for each target person such as a learner.

It is a block diagram of the network which shows schematic the learning continuation system which concerns on one Embodiment of this invention. It is a functional block diagram which shows schematic the internal structure of the server of FIG. It is a flowchart explaining the control of information acquisition executed by the control part of FIG. It is a flowchart explaining the control of withdrawal index calculation executed by the control part of FIG. It is a flowchart explaining the control of the evaluation index calculation executed by the control part of FIG. It is a flowchart explaining the control of the measure decision executed by the control part of FIG.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a network configuration diagram schematically showing a learning continuation system according to an embodiment of the present invention. The learning continuation system 10 includes a network 11, a learner terminal 12, an operator terminal 13, a server 14, a memory 15, and the like. The learning continuation system 10 is used by an operator who provides learners with educational services (causatives to be provided) such as correspondence education, learning courses, and regular examinations.

The network 11 includes at least one of wired and wireless or connects a plurality of learner terminals 12, a plurality of operator terminals 13, and a server 14. The network 11 transmits information and commands between the learner terminal 12, the operator terminal 13, and the server 14.

The learner terminal 12 is an information processing device such as a tablet terminal, a smartphone, or a personal computer used by the learner (target person). The learner terminal 12 acquires the video of the lecture, the question group of a predetermined combination, the question group that fluctuates based on the answer record, the menu screen, and the like provided from the server 14 via the network 11. indicate. The menu screen includes points given to the learner by the operator based on the result of the answer to the question.

In addition, the learner terminal 12 detects the user's input to the displayed image and executes a function corresponding to the detected input. For example, when the learner terminal 12 detects inputs such as pause, rewind, fast forward, and stop in the video of the course, the video is paused, rewound, fast forward, and stopped, respectively. Further, when the learner terminal 12 detects the input of the answer while displaying the question group, the learner terminal 12 transmits the answer to the server 14 and displays the correct answer and the explanation. In addition, when the learner terminal 12 detects input such as a question about learning, a request and question about the operator, a use of granted points, a question about life, and a question about the course while displaying the menu screen, the input contents are sent to the server. Notify 14

The operator terminal 13 is an information processing device such as a tablet terminal, a smartphone, or a personal computer used by the operator. The operator terminal 13 displays an image for inputting various information necessary for providing the educational service, detects the input in the input image, and notifies the server 14. For example, the operator terminal 13 detects an input of either the withdrawn or ongoing status of the current learner and the past learner. In addition, the operator terminal 13 detects the input of the measures executed for the current learner and the past learner. In addition, the operator terminal 13 detects the input of information regarding the learning history and behavioral characteristics of each learner, and notifies the server 14. The operator terminal 13 requests the user to input each item of "when", "who", "what", and "what" with respect to the information related to the learning history and the behavior identification.

The measures taken are, for example, calling out to either the learner or the guardian, "Are you busy with club activities?", "How are you doing in school math classes these days?", "I don't know in school classes." Is there any place? ”,“ I'm a little worried about the answer to English grammar, but would you like to try learning this part again using similar questions? ”,“ What are you worried about in your life? Are there any questions, such as "Are you thinking about your future dreams and career paths?", And changes in the educational services provided.

Information on the learning history includes, for example, scoring results and comments by the grader, scoring results for the answers on the learner terminal 12, grades in correspondence education and regular examinations based on mail, digestion status of homework provided to the learner, and learning. It is the state of understanding of each learner's course by the instructor of the course, and the state of the learner's efforts such as repeated viewing and answering for the educational role such as the course and problem group provided to the learner.

Information on behavioral characteristics includes, for example, the attendance status of each learner in the learning course, the absolute or relative entry / exit time of each learner in the learning course (note that the relative entry / exit time is the learning course. The difference between the entry time and exit time for each of the start time and end time of Answers to the questionnaire, the status of taking learning courses recommended for each learner, the status of attendance at briefing sessions such as parents' associations, the number of consultations from each learner and the content of consultation, the ability of each learner to concentrate on the course , Living conditions such as extracurricular activities of each learner, consideration of reducing the number of courses taken by each learner, presence or absence of worries about the course of each learner, troubles between operators other than the educational services provided and each learner, Depending on the degree of noise of the learning course that each learner takes, the change in the responsibility of the operator for each learner, the course status of each learner's relatives, the awareness of each learner for the exam, and the awareness of each learner for the exam. be.

Further, the operator terminal 13 outputs various information based on the management and processing of the server 14 via, for example, a monitor and a printer. For example, the operator terminal 13 outputs an unconfirmed answer acquired from the server 14. Further, the operator terminal 13 outputs a predetermined statistical value acquired from the server 14. Further, the operator terminal 13 outputs a measure to be taken by the operator, which is acquired from the server 14.

The server 14 is, for example, an information processing device such as a server device used by an operator. As shown in FIG. 2, the server 14 has a communication unit 16 and a control unit 17. The communication unit 16 is connected to the network 11 and the control unit 17, and transmits information and commands between the network 11 and the control unit 17. The control unit 17 is connected to the memory 15 and stores information in the memory 15 and reads information from the memory 15 as needed.

The control unit 17 provides and manages educational services.

For example, the control unit 17 sends a video of the lecture to the learner terminal 12 based on the command of the learner terminal 12 or the command from the operator terminal 13 or at a predetermined time. Further, the control unit 17 acquires the inputs of pause, rewind, fast forward, and stop in the learner terminal 12 that sent the video of the lecture together with the input detection time in the video. Further, the control unit 17 associates the acquired pause, rewind, fast forward, and stop inputs with the input detection timing in the video, and stores the acquired inputs in the memory 15 for each learner.

Further, for example, the control unit 17 sends a problem group to the learner terminal 12 based on a command from the learner terminal 12 or a command from the operator terminal 13 or at a predetermined time. Further, the control unit 17 stores the answer to the question group acquired from the learner terminal 12 in the memory 15, assigns a grader to score the answer, and sends the answer to the operator terminal 13 used by the grader as an unconfirmed answer. Alternatively, the control unit 17 may score based on the correct answer read from the memory 15. Further, the control unit 17 sends a correct answer and an explanation to the learner terminal 12. Further, the control unit 17 adds and stores the points for giving the learner the privilege according to the numerical value to the points stored in the memory 15 based on the scoring result. Further, the control unit 17 stores the scoring results obtained from the operator terminal 13, the results in the correspondence education based on mail and the regular examination, and the scoring results of the answers scored by the control unit 17 itself in the memory 15 for each learner. Remember.

Further, for example, the control unit 17 calculates a predetermined statistical value regarding the scoring results of a plurality of learners based on the scoring results. Further, the control unit 17 notifies the operator terminal 13 of the calculated predetermined statistical value based on the command of the operator terminal 13. Further, the control unit 17 stores the calculated predetermined statistical value in the memory 15 for each learner.

Further, for example, the control unit 17 sends a menu screen or the like to the learner terminal 12 based on the command of the learner terminal 12. Further, the control unit 17 asks each learner a question about learning, a request and a question about the operator, a question about the use of the given points, a question about life, and a question about the course, which are acquired from the learner terminal 12 and are input on the menu screen. It is stored in the memory 15.

Further, for example, the control unit 17 stores in the memory 15 the status of withdrawal or continuation for each learner acquired from the operator terminal 13. Further, the control unit 17 stores in the memory 15 the measures executed for each learner acquired from the operator terminal 13. In addition, the control unit 17 provides scoring results and comments by the grader, scoring results for the answers on the learner terminal 12, information on learning history other than grades in correspondence education based on mail and regular examinations, and information on behavioral characteristics. , Stored in memory 15.

Further, the control unit 17 processes the stored information based on the learning history and the behavioral characteristics stored in the memory 15. For example, the control unit 17 estimates the repetitive viewing points in the lecture based on the rewind input for the video lecture, and stores it in the memory 15. Further, the control unit 17 estimates a problem whose answer tends to be prolonged based on the acquisition interval of the answer to the problem group sent to the learner terminal 12, and stores it in the memory 15. Further, the control unit 17 stores the transition of the grade for each learner in the memory 15 based on the scoring result and the grade of the test. Further, the control unit 17 totals the number of questions in a predetermined period or a single course for each learner and stores them in the memory 15. Further, the control unit 17 analyzes the content of the question for each learner, determines whether or not the question is a negative question, and stores the determination result in the memory 15. A negative question is a question that does not lead to an understanding of the learning content, and is a question on the premise that the problem or the like cannot be solved or understood. Further, the control unit 17 determines whether or not substantially all the points given to each learner are used, and stores the determination result in the memory 15.

Further, the control unit 17 may change the determination as to whether or not the question is the above-mentioned negative question for each learner. For example, the control unit 17 asks, "When is the answer submitted?", "How long does it take to solve the problem?", "When to ask a question, is it in the middle of solving the problem or when the answer is confirmed?" The standard state of each learner is analyzed based on about 50 pieces of information such as "What is the reaction to the notification from the instructor?" The analysis of the standard state may be performed by, for example, an evaluation based on a behavior model, a combination, or a conventionally known evaluation method using a decision tree. The control unit 17 may determine whether or not the question is negative based on the learner's standard state.

Further, the control unit 17 predicts that the learner will withdraw based on the behavioral characteristics and learning history of the learner who has been withdrawn and the information processed from the behavioral characteristics and the learning history, which are read from the memory 15. Build a model.

In the present embodiment, the construction of the predictive model includes, for example, detection of correlation and estimation of behavioral characteristics and learning history leading to learner withdrawal based on the detected correlation. The correlation is the correlation between the behavioral characteristics and learning history of the learner who has withdrawn, and the information processed based on the behavioral characteristics and learning history, and the withdrawal status.

To exemplify the detection of correlation in more detail, the detection of correlation is performed by analysis of big data. For example, the correlation coefficient is calculated and a predetermined threshold value or all behavioral characteristics and all learning history are detected. It may be a comparison with a statistically determined threshold value such as the upper x1% (x1 is a positive real number).

Further, the control unit 17 estimates that the behavioral characteristics and the learning history in which the correlation with the withdrawn situation is the behavioral characteristics and the learning history leading to the withdrawal of the learner are added to the behavioral characteristics and the learning history. Build a predictive model of withdrawal based on.

Further, the control unit 17 evaluates the measures taken for the target person based on the prediction model, that is, the behavioral characteristics estimated to lead to the withdrawal of the learner, the learning history, and the like. Further, the control unit 17 determines the measures to be taken by the operator based on the evaluation result. Further, the control unit 17 notifies the operator terminal 13 of the determined measures to be taken.

To exemplify in more detail the estimation of the behavioral characteristics and learning history leading to the withdrawal of the learner, the control unit 17 is processed from the behavioral characteristics and learning history of the learner, which is an ongoing situation, and the behavioral characteristics and learning history. The withdrawal index is calculated for each learner by reading out the information obtained and the correlation with the withdrawal status and analyzing the big data. The control unit 17 performs big data analysis based on, for example, a support vector machine and regression analysis. The withdrawal index is an index corresponding to the possibility of the learner's withdrawal. Further, the control unit 17 stores the calculated withdrawal index in the memory 15.

To exemplify in more detail the evaluation of the measures taken for the learner based on the prediction model, the control unit 17 has withdrawn from the past in the combination of a plurality of withdrawal indexes calculated at different times and stored in the memory 15. Learners whose index is greater than the threshold and whose latest withdrawal index is less than or equal to the threshold are extracted from a plurality of learners. Further, the control unit 17 counts the extracted number of learners as a population. Further, the control unit 17 reads the measures executed for the extracted learners from the memory 15.

Further, the control unit 17 calculates an evaluation index for each measure executed after reading. The evaluation index is an index showing the effectiveness of the implemented measures. The calculation of the evaluation index is based on the ratio of the number of learners counted as a population to the number of learners who have implemented the measures. In addition, the evaluation index is calculated for each measure and for each behavioral characteristic and learning history of the learner who carried out the implemented measure. In other words, the evaluation index is not simply calculated for each measure taken, but comprehensively evaluates measures for learners with specific behavioral characteristics and a specific learning history. Further, the control unit 17 stores the calculated evaluation index in the memory 15.

To exemplify in more detail the presentation of measures to be taken based on the evaluation result, the control unit 17 executed the learner whose latest withdrawal index is larger than the threshold for the learner's behavioral characteristics and learning history. Among the measures, the measure with the highest evaluation index is decided as the measure to be taken. The control unit 17 may determine the measure that has the largest evaluation index among the implemented measures as the measure to be taken, or has executed a plurality of measures such as the top x2 (x2 is an integer of 2 or more). The measure may be decided as the measure to be taken. Further, the control unit 17 notifies the operator terminal 13 of the determined measures to be taken.

Further, when the learner's approach to the educational causative provided to the learner, which is a part of the learning history, the control unit 17 presents the learner by notifying the operator terminal 13 as a measure to take a specific measure. You may. For example, the control unit 17 calculates the above-mentioned interoperability based on the occurrence rate of the learner's viewing interruption and repeated viewing, which is the status of withdrawal, in the provided course. Further, when the interactivity exceeds the threshold value set to an appropriate value of, for example, 0.3 or more, the control unit 17 presents the improvement of the educational causative based on the learner's efforts as a measure to be taken.

Next, in the present embodiment, the control of information acquisition executed by the control unit 17 will be described with reference to the flowchart of FIG. When the control unit 17 acquires the information from the network 11, the control unit 17 starts the control of the information acquisition.

In step S100, the control unit 17 stores the acquired information in the memory 15 for each learner. Upon storage, the process proceeds to step S101.

In step S101, the control unit 17 determines whether or not the information stored in the memory 15 in step S100 is processable information together with the information already stored in the memory 15 as needed. When processable, the process proceeds to step S102. When processing is not possible, the process skips steps S102 and S103 and ends the control of hand information acquisition.

In step S102, the control unit 17 reads the information stored in the memory 15 in step S100 and, if necessary, the information already stored in the memory 15 from the memory 15. Further, the control unit 17 processes the read information. After processing, the process proceeds to step S103.

In step S103, the control unit 17 stores the information processed in step S102 in the memory 15. When stored in the memory 15, the control of information acquisition ends.

Next, in the present embodiment, the control of the withdrawal index calculation executed by the control unit 17 will be described with reference to the flowchart of FIG. The control unit 17 starts controlling the withdrawal index calculation periodically or periodically.

In step S200, the control unit 17 reads out the behavioral characteristics and the learning history of the learner who has been withdrawn from the memory 15. Upon reading from memory 15, the process proceeds to step S201.

In step S201, the control unit 17 detects the correlation between the behavioral characteristics and the learning history read from the memory 15 in step S200 and the situation of withdrawal. Further, the control unit 17 calculates the withdrawal index for each learner based on the behavioral characteristics in which the correlation is detected and the learning history. After calculating the withdrawal index, the process proceeds to step S202.

In step S202, the control unit 17 stores the withdrawal index calculated in step S201 in the memory 15 for each learner. When stored in the memory 15, the control of the withdrawal index calculation ends.

Next, in the present embodiment, the control of the evaluation index calculation executed by the control unit 17 will be described with reference to the flowchart of FIG. The control unit 17 starts controlling the evaluation index calculation periodically or periodically.

In step S300, the control unit 17 extracts learners whose past withdrawal index is larger than the threshold value and whose latest withdrawal index is less than or equal to the threshold value. Further, the control unit 17 counts the number of extracted learners. Further, the control unit 17 reads the measures executed for the extracted learners from the memory 15. Upon reading from memory 15, the process proceeds to step S301.

In step S301, the control unit 17 calculates an evaluation index for each measure executed read in step S300. After calculating the evaluation index, the process proceeds to step S302.

In step S302, the control unit 17 stores the evaluation index calculated in step S301 in the memory 15. When stored in the memory 15, the control of the evaluation index calculation ends.

Next, in the present embodiment, the control of the measure determination executed by the control unit 17 will be described with reference to the flowchart of FIG. The control unit 17 starts controlling the policy decision based on the request command from the operator terminal 13.

In step S400, the control unit 17 reads from the memory 15 the behavioral characteristics and the learning history of the learner whose latest withdrawal index is larger than the threshold value. Upon reading, the process proceeds to step S401.

In step S401, the control unit 17 reads from the memory 15 the measure having the higher evaluation index for the behavioral characteristics and the learning history read in step S400 among the executed measures. Upon reading, the process proceeds to step S402.

In step S402, the control unit 17 notifies the operator terminal 13 of the measures read in step S401. When the operator terminal 13 is notified, the control of the policy decision is terminated.

According to the learning continuation system 10 of the present embodiment having the above configuration, a predictive model leading to the withdrawal of the learner is constructed based on the behavioral characteristics and the learning history of the learner who has withdrawn, and based on the predictive model. It is possible to evaluate the measures taken and present the measures to be taken based on the evaluation results. Therefore, it can provide appropriate measures that can encourage the continuation of educational services for each learner.

Further, according to the learning continuation system 10 of the present embodiment, it can be presented as a measure for improving the educational service based on the learner's approach to the educational service provided to the learner. Therefore, the improvement of the educational service can reduce the learner's dissatisfaction, and as a result, the learner's continuation of the educational service can be promoted.

Further, according to the learning continuation system 10 of the present embodiment, the measures taken for the learner whose withdrawal index is high in the past and later whose withdrawal index is low are evaluated. The measure may reduce the likelihood of withdrawal. Furthermore, it is possible for the operator to take appropriate measures for learners who may withdraw after evaluating the measures and based on the evaluation results. Therefore, it can encourage learners to continue their educational services.

Although the present invention has been described based on the drawings and embodiments, it should be noted that those skilled in the art can easily make various modifications and modifications based on the present disclosure. Therefore, it should be noted that these modifications and modifications are included in the scope of the present invention.

For example, in the present embodiment, various behavioral characteristics and various learning histories are configured to detect the correlation between the withdrawal status and various behavioral characteristics and various learnings for each period such as monthly. The configuration may be such that the correlation between the history and the situation of withdrawal is detected, and the measures to be taken based on the behavioral characteristics and the learning history in which the correlation is detected are presented for each period. Since the factors by which the learner considers withdrawal from the educational service may change depending on the time, the measures to be presented can be optimized by the above-mentioned configuration.

10 Learning continuation system 11 Network 12 Learner terminal 13 Operator terminal 14 Server 15 Memory 16 Communication unit 17 Control unit

Claims (8)

A memory that stores behavioral characteristics and learning history for each of a plurality of subjects, as well as withdrawal or ongoing situations , and measures taken for each subject.
To build a predictive model leading to withdrawal of the subject of the withdrawal already based on the behavioral characteristics and the history of learning before Symbol subject of the withdrawal already, withdrawal index for behavioral characteristics and learning history on the basis of the predictive model Withdrawal from the time of calculation of the past withdrawal index for the ongoing subject whose withdrawal index is larger than the threshold value and the subsequent withdrawal index is equal to or less than the threshold value. A learning continuation system characterized in that it is equipped with a control unit that reads the measures executed by the time the index is calculated from the memory , evaluates the measures, and presents the measures to be taken based on the evaluation results. The learning continuation system according to claim 1.
The learning history, said # more than leave includes efforts status of the subject for the causative providing the subject in the continuation,
The learning continuation system is characterized in that the control unit presents as a measure to improve the causative based on the action situation when the correlation between the action situation and the withdrawn situation exceeds the threshold value. .. The learning continuation system according to claim 1 or 2.
Construction of the prediction model, the withdrawal already situations, as well as the detection of the correlation of the behavioral characteristics and the history of learning before Symbol subject's dehydration Shisasumi, said detected withdrawal the already said, based on the correlation A learning continuation system comprising the behavioral characteristics leading to the withdrawal of a subject and the estimation of the learning history. The learning continuation system according to claim 3.
The estimation of the behavior characteristics and the learning history leading to withdrawal of withdrawal already subjects, the plurality of subjects for each, based on the behavioral characteristics and the learning history is detected correlation with the state of the withdrawal already learning continuation system comprising the calculation of the withdrawal index. The learning continuation system according to claim 4 .
Prior Symbol controller, in the evaluation of the measures run against the subject of the ongoing based on the prediction model, the withdrawal exponent is stored in the memory, past the withdrawal index than the latest read out from said memory Is larger than the threshold value and the latest withdrawal index is equal to or less than the threshold value. For each of the behavioral characteristics and the learning history in which the correlation with the read-out and withdrawal status is detected, the number of the target persons who have performed the implemented measure is relative to the counted number of target persons. A learning continuation system characterized by calculating an evaluation index evaluated based on a ratio. The learning continuation system according to claim 5.
In presenting measures to be taken based on the evaluation result, the control unit gives the target person whose latest withdrawal index is larger than the threshold value the evaluation index for the behavioral characteristics and the learning history of the target person. A learning continuation system characterized by presenting the higher-ranked measures taken as the measures to be taken. The learning continuation system according to any one of claims 3 to 6.
The memory stores the behavioral characteristics and the learning history, and the withdrawn or ongoing situation at each time period.
The learning continuation system is characterized in that the control unit detects the correlation at each time period and presents the measures to be taken at each time period based on the behavioral characteristics and the learning history in which the correlation is detected. .. It is a learning continuation method that is executed by a computer.
A memory step that memorizes the behavioral characteristics and learning history of each of a plurality of subjects, the status of withdrawal or ongoing , and the measures taken for each subject.
The construction steps based on the behavioral characteristics and the history of learning before Symbol subject of the withdrawal already, to build a prediction model leading to withdrawal of the subject of the withdrawal already,
Based on the prediction model, a calculation step for calculating the withdrawal index for behavioral characteristics and learning history at different times, and
From the time when the past withdrawal index is calculated to the time when the subsequent withdrawal index is calculated for the ongoing subject whose withdrawal index calculated in the past is larger than the threshold value and the subsequent withdrawal index is equal to or less than the threshold value. The evaluation step to read the measures taken in the memory from the memory and evaluate the measures,
A learning continuation method characterized by having a presentation step that presents measures to be taken based on the evaluation result.

Images