JP7480441B1 - QUESTION CREATION SUPPORT DEVICE, QUESTION CREATION SUPPORT METHOD, AND PROGRAM - Google Patents

Abstract

[Problem] In the past, the number of questions to obtain a total score was large, placing a heavy burden on members. [Solution] A question creation support device 1 is equipped with a correlation acquisition unit 132 that acquires correlation information regarding the correlation between items corresponding to one or more questions and the total score using a member score that includes an answer score based on the member's answer to each of N questions (N is a natural number of 3 or more) for each of two or more members and the member's total score, a decision unit 133 that performs one or more processes of a deletion decision process that determines a question to be deleted from the N questions, or an integration decision process that determines two or more questions to be integrated into one of the N questions, using the correlation information for each item, and acquires decision information regarding the decision, and an output unit 14 that outputs the decision information, so that an appropriate number of questions can be proposed, thereby reducing the burden on members when obtaining a total score. [Selected Figure] Figure 1

Description

The present invention relates to a question creation support device that supports the creation of questions that make up a questionnaire.

A conventional personnel management system has been developed that aims to provide a technology for analyzing the relationship between personnel and labor information and engagement, and is characterized by comprising a database for managing employee personnel and labor information, a question sending unit for sending work-related questions to the employees, an answer receiving unit for receiving answers to the questions from the employees, an index calculation unit for calculating the employee's engagement based on the answers, and an analysis unit for analyzing the relationship between the engagement and the personnel and labor information contained in the database (see Patent Document 1).

Patent No. 6845595

However, with conventional technology, the number of questions required to obtain an overall score for engagement, etc., was large, placing a heavy burden on members.

The question creation support device of the first invention is a question creation support device that includes a correlation acquisition unit that acquires correlation information regarding the correlation between items corresponding to one or more questions and a total score using a member score that has an answer score based on the member's answer to each of N questions (N is a natural number equal to or greater than 3) for each of two or more members and a total score for the member, a decision unit that performs one or more of a deletion decision process that determines which of the N questions to delete, or a merge decision process that determines which of the N questions to merge into one of the N questions, using the correlation information for each item, and acquires decision information regarding the decision, and an output unit that outputs the decision information.

This configuration allows an appropriate number of questions to be proposed, thereby reducing the burden on members when obtaining an overall score.

The question creation support device of the second invention, compared to the first invention, further includes a member score acquisition unit that acquires, for each of the two or more members, a member score having an answer score based on the member's answer to each of the N questions and a total score for the member, and the two or more member scores used by the correlation acquisition unit are the two or more member scores acquired by the member score acquisition unit.

This configuration allows an appropriate number of questions to be proposed, thereby reducing the burden on members when obtaining an overall score.

The question creation support device of the third invention is a question creation support device according to the first or second invention, in which each of the N questions corresponds to a classification identifier that identifies one of two or more classifications of questions, and the determination unit performs one or more of a deletion determination process that determines which questions to delete from the M (3<=M<N) questions that correspond to the classification identifier that identifies the classification, or a merger determination process that determines which two or more questions to merge from the M questions, for each of the two or more classifications, and acquires the determination information.

This configuration ensures a balance in the number of questions in each category, resulting in a well-balanced and appropriate set of questions when obtaining an overall score.

The question creation support device of the fourth invention is a question creation support device according to the third invention, in which the decision unit performs one or more of a deletion decision process that decides which questions to delete from among M questions corresponding to the classification identifiers so that the number of questions corresponding to each of the two or more classification identifiers is the same or the variation is below a threshold, or a merger decision process that decides which two or more questions to merge into one of the M questions and obtains decision information regarding the decision.

This configuration ensures a balance in the number of questions in each category, resulting in a more balanced and appropriate set of questions when obtaining an overall score.

The question creation support device of the fifth invention is a question creation support device according to the third or fourth invention, in which an answer score corresponding to at least one question corresponds to one of the classification identifiers, and a process identifier that specifies a process to be performed by the determination unit corresponds to each of two or more correlation pattern identifiers that specify a pattern of correlation information of M answer scores corresponding to the same classification identifier, and the process specified by the process identifier is either a deletion determination process or an integration determination process, and the determination unit obtains, for each of the two or more classifications, a correlation pattern identifier that specifies a pattern of correlation information of M answer scores that corresponds to the classification identifier that specifies the classification, obtains a process identifier that pairs with the correlation pattern identifier, performs the process specified by the process identifier, and obtains the determination information.

This configuration allows an appropriate number of questions to be proposed, thereby reducing the burden on members when obtaining an overall score.

The question creation support device of the sixth invention is a question creation support device according to any one of the second to fifth inventions, in which the member scores correspond to one or more organizational attribute values, the member score acquisition unit acquires the member scores of two or more members who satisfy an organizational condition using one or more organizational attribute values, and the correlation acquisition unit acquires correlation information using the two or more member scores acquired by the member score acquisition unit.

This configuration allows us to propose an appropriate set of questions for each organizational condition.

The question creation support device of the seventh invention is a question creation support device according to the first or second invention, in which each of the N questions corresponds to one of two or more classification identifiers, and each of the three or more answer scores corresponding to at least one question corresponds to one of the classification identifiers, the correlation acquisition unit acquires correlation information between the two or more answer scores corresponding to different classification identifiers, and the determination unit performs one or more of a deletion determination process for determining a question to be deleted that corresponds to at least one answer score of the two or more answer scores using the correlation information between the two or more answer scores, or a merger determination process for determining two or more questions to be merged into one that corresponds to at least one answer score of the two or more answer scores, to acquire the determination information.

This configuration allows an appropriate number of questions to be proposed, thereby reducing the burden on members when obtaining an overall score.

The question creation support device of the eighth invention is a question creation support device according to any one of the first to seventh inventions, in which the N questions have two or more questions and are X (X>=3) sets of question sets corresponding to the items, and the determination unit uses correlation information of two or more answer scores to perform one or more of a deletion determination process for determining a question set to be deleted from the X sets of question sets, or a merge determination process for determining two or more question sets to be merged from the X sets of question sets, and acquires the determination information.

This configuration allows an appropriate number of questions to be proposed, thereby reducing the burden on members when obtaining an overall score.

The question creation support device of the ninth invention is a question creation support device in which, compared to the eighth invention, the question set has two questions asking about the expectation and satisfaction with the item, and the overall score is a score obtained using item scores obtained using two answers, the answer about the expectation with respect to the item and the answer about the satisfaction with the item.

This configuration reduces the burden on members when obtaining a total score based on a questionnaire asking about their expectations and satisfaction with items.

The question creation support device of the tenth invention is a question creation support device according to the ninth invention, in which the correlation acquisition unit acquires, for each item, first correlation information relating to the correlation between the answer score based on the response regarding the expectation for the item and the overall score, acquires second correlation information relating to the correlation between the answer score based on the response regarding the satisfaction for the item and the overall score, acquires correlation information based on the first correlation information and the second correlation information, and the determination unit uses the correlation information for each item to perform one or more of a deletion determination process for determining which question sets to delete from among the X question sets, or a merger determination process for determining which two or more question sets to merge from among the X question sets, and acquires the determination information.

This configuration reduces the burden on members when obtaining a total score based on a questionnaire asking about their expectations and satisfaction with items.

The question creation support device of the present invention can propose an appropriate number of questions, thereby reducing the burden on members when obtaining a total score.

Block diagram of a question creation support device 1 according to the first embodiment. A flowchart for explaining an example of the operation of the simultaneous question creation support device 1 11 is a flowchart illustrating a first example of the determination process. A flowchart illustrating an example of the classification and aggregation process. A flowchart illustrating an example of the determination information acquisition process. 11 is a flowchart illustrating a second example of the determination process. A flowchart illustrating an example of the aggregation process. The figure shows the item information management table. A diagram showing the response information A diagram showing the individual score table The pattern management table is shown in FIG. A chart showing the member score management table A diagram showing examples of items to be aggregated Overview of the computer system Block diagram of the computer system

Below, an embodiment of a question creation support device and the like will be described with reference to the drawings. Note that components with the same reference numerals in the embodiments perform similar operations, and therefore repeated explanations may be omitted.

(Embodiment 1)
In this embodiment, a question creation support device is described that obtains correlation information regarding the correlation between items for multiple questions and a total score, and supports reducing the number of questions by deleting some questions from the multiple questions and/or integrating some questions. Note that the total score is, for example, a score called an engagement score.

In addition, in this embodiment, a question creation support device is described that obtains correlation information between items and total scores for each of two or more categories, and supports reducing the number of questions by deleting some questions and/or merging some questions.

In addition, in this embodiment, a question creation support device is described that acquires correlation information between items of different classifications and supports reducing the number of questions by deleting some questions and/or merging some questions.

In addition, in this embodiment, a question creation support device is described that supports the determination of questions so that the number of questions in each of two or more categories is the same or has less variation.

In this embodiment, a question creation support device is described that determines a pattern of correlation information for items in a classification, and provides support for reducing the number of questions by deleting some questions and/or merging some questions according to the pattern.

Furthermore, in this embodiment, a question creation support device that provides support for aggregating questions for each of one or more conditions will be described.

In this specification, information X being associated with information Y means that information Y can be obtained from information X, or information X can be obtained from information Y, and the method of association is not important. Information X and information Y may be linked, may exist in the same buffer, information X may be included in information Y, or information Y may be included in information X, etc.

Figure 1 is a block diagram of a question creation support device 1 in this embodiment. The question creation support device 1 includes a storage unit 11, a reception unit 12, a processing unit 13, and an output unit 14. The storage unit 11 includes an item information storage unit 111, an answer information storage unit 112, an individual score table storage unit 113, an organizational information storage unit 114, a pattern management unit 115, and a member score management unit 116. The reception unit 12 includes a condition reception unit 121. The processing unit 13 includes a member score acquisition unit 131, a correlation acquisition unit 132, and a determination unit 133. The member score acquisition unit 131 includes an answer score acquisition means 1311 and a total score acquisition means 1312.

Various types of information are stored in the storage unit 11 constituting the question creation support device 1. The various types of information include, for example, item information (described later), answer information (described later), individual score tables (described later), organizational information (described later), pattern information (described later), and member scores (described later). The individual score tables may also be referred to as engagement score tables, etc.

The item information storage unit 111 stores two or more pieces of item information. The item information is information about items. The item information is, for example, information about items of an organization. The items may be questions about an organization. Also, the items may correspond to one or two or more questions. The two or more items include, for example, comprehensive items related to comprehensive matters of the organization and individual items that are individual items of the organization. Note that the comprehensive items are usually items for questions with a high level of abstraction. Also, the individual items are usually items for questions with a lower level of abstraction (more specific questions). Also, the two or more pieces of item information in the item information storage unit 111 include, for example, item information for four comprehensive items and item information for 64 individual items. The two or more pieces of item information in the item information storage unit 111 may be, for example, item information for 64 individual items. Also, the items correspond to, for example, any of two or more classifications. The classification may be called a target or a factor. Also, the classification may be called, for example, matters related to the organization. Furthermore, the item information has, for example, an item identifier that identifies an item, and question information. Furthermore, the item identifier is, for example, an ID, an item name, etc. The item identifier may be the question information itself. The question information is information that indicates a question. The question information is simply called a "question" as appropriate. Note that a question is usually a question in a questionnaire. A question may also be called a question. It is preferable that the question is a question about the organization, but this is not essential.

In addition, in this specification, the content of the questions does not matter. In addition, in this specification, the subject of the questions does not have to be the members of the organization. Members may also be called users. However, in this specification, it is preferable that each of the two or more items corresponds to a question group having two or more questions. The question group for an item is, for example, two or three or more questions including a question asking about the expectation level for a certain item and a question asking about the satisfaction level. It goes without saying that the expectation level may be replaced with other terms having the same, similar or related meanings, such as the importance level or the degree of importance. It goes without saying that the satisfaction level may be replaced with other terms having the same, similar or related meanings, such as the evaluation value, the degree of achievement, the degree of fulfillment, etc.

The answer information storage unit 112 stores answer information for two or more members. The answer information is information that indicates the answer to the question for the member. Note that the member may simply be referred to as a user.

The response information is associated with, for example, an organization identifier that identifies the organization. The organization identifier is an organization name, an ID that identifies the organization, etc. The response information is associated with, for example, an organization attribute value that is one or more attribute values of the organization. The organization attribute value is, for example, an industry identifier that indicates the industry of the organization (for example, bank, apparel, manufacturer, etc.), a scale identifier that classifies the size of the organization (for example, large company, small and medium-sized enterprise, micro-enterprise, sole proprietorship, etc.), information that indicates the category of the organization based on the management status of the organization (for example, a profitable company, a loss-making company, etc.), a region identifier that indicates the region of the head office, a department identifier that identifies a department within the organization (for example, human resources, accounting, research institute, technical department, sales department, manufacturing department, etc.), a listing identifier that indicates whether the organization is listed on the stock exchange, a stage identifier that indicates the stage of the organization (for example, the founding period, the expansion period, the diversification period, the revitalization period, etc.), a business model identifier that indicates the business model (for example, an innovator who conducts multiple types of businesses, a professional who conducts a few types of businesses such as one and is dependent on people, an operator who conducts a few types of businesses such as one and is dependent on systems, etc.), etc. The industries can be broad categories such as finance, manufacturers, trading companies, and services, or smaller categories such as banks, securities, electrical manufacturers, food manufacturers, and machinery manufacturers. It goes without saying that there is no restriction on how industries are divided.

A collection of member answer information for an organization is called organizational answer information. Organization answer information has one or more member answer information corresponding to one or more members. Member answer information is information including the results of a member's answer to a question for an item. Member answer information has two or more item answer information. Member answer information usually has item answer information for the number of items. Item answer information has an item identifier and answer information. Answer information is information related to the answer to the question. Answer information has, for example, satisfaction information. Note that satisfaction information may simply be called "satisfaction". Satisfaction information is information indicating the answer regarding the member's degree of satisfaction with an item. Satisfaction information is, for example, information that specifies the satisfaction with an item. Satisfaction information is classified into two or more classes. Satisfaction information may be, for example, any natural number from 1 to 5. However, satisfaction information may be, for example, an evaluation value with a rank or order, such as A, B, C, or any natural number from 1 to 100. Also, answer information has, for example, satisfaction information and expectation information. The expectation information is information that indicates the member's response regarding the degree of expectation for an item. The expectation information may simply be called "expectation." The expectation information is, for example, information that specifies the degree of expectation for an item. The expectation information is classified into two or more classes. The expectation information may be, for example, any natural number from 1 to 5. However, the expectation information may be, for example, an evaluation value with a rank or order, such as A, B, C, or any natural number from 1 to 100. The contents of the items and questions are not important. Furthermore, the members may be, for example, company employees, school staff, government staff, etc., but may also include company executives, etc. The members may be part-time workers.

In addition, examples of comprehensive items include company satisfaction, which indicates the degree of satisfaction with the company; job satisfaction, which indicates the degree of satisfaction with the job; supervisor satisfaction, which indicates the degree of satisfaction with the supervisor; and workplace satisfaction, which indicates the degree of satisfaction with the workplace. In addition, examples of individual items include the company's business advantages, the dissemination and communication of strategic goals, an overall sense of solidarity, and the appropriateness of evaluations and wages.

The individual score table storage unit 113 stores an individual score table. The individual score table here is information for determining item scores when satisfaction information and expectation information are given. The individual score table here is, for example, a table having an axis of satisfaction information and an axis of expectation information, in which an individual score is entered in each cell of the table, and in which item scores are determined when satisfaction information and expectation information are determined. It is preferable that such an individual score table is a table in which the greater the satisfaction indicated by the satisfaction information, the higher the item score obtained, and the smaller the expectation indicated by the expectation information, the higher the item score obtained.

An item score is a score obtained based on the answers to one or more questions. An item score is usually a score based on the answers to two or more questions in a question set. However, an item score may also be a score based on the answer to one question. A question set is a set of two or more questions. A question set is, for example, a set of a question asking about expectations for an item and a question asking about satisfaction with the item. It is preferable that an item score is a score obtained using two answers, an answer about expectations for an item and an answer about satisfaction with the item. An item score is, for example, a score paired with two answers, an answer about expectations for an item and an answer about satisfaction with the item, and is a score obtained from an individual score table.

The item score may be determined by an arithmetic formula with satisfaction information and expectation information as parameters. Preferably, the arithmetic formula is an increasing function with satisfaction information as a parameter, and a decreasing function with expectation information as a parameter.

In addition, multiple sets of satisfaction information, expectation information, and item scores may be learned by machine learning, and the acquired learning information may be used to determine the item scores. In such a case, the satisfaction information, expectation information, and learning information are applied to acquire the item scores by machine learning. Note that the machine learning used here may be, for example, SVR, deep learning, decision trees, random forests, etc. However, any machine learning algorithm may be used.

Two or more pieces of organizational information are stored in the organizational information storage unit 114. The organizational information is information related to an organization. The organizational information has an organization identifier and one or more organization attribute values. It is preferable that the organizational information has an organization identifier, an organization attribute value, and an overall score that is a comprehensive score for the organization. The overall score here may be an absolute overall score that does not depend on the organization attribute value.

An overall score is a score obtained using two or more answer scores. An overall score is, for example, a score obtained using two or more item scores. A score obtained using item scores is typically a score obtained using answer scores. An overall score is, for example, an overall score for one member. An overall score may also be an overall score for an organization based on the overall scores of two or more members.

The pattern management unit 115 stores two or more pieces of pattern information. The pattern information is information that indicates the correspondence between a correlation pattern identifier and a processing identifier. The pattern information includes, for example, correlation pattern information and processing information.

The correlation pattern identifier is information that identifies a pattern related to correlation information of two or more answer scores. It can be said that the correlation pattern identifier is information that identifies correlation pattern information. The pattern related to correlation information of answer scores may be a pattern related to correlation information of item scores. The pattern may also be referred to as a condition, a trend, etc. The correlation pattern information is information that specifies a pattern of correlation information of two or more items. The correlation pattern information is, for example, information that specifies the relationship between correlation information of two or more items, or information that specifies a threshold value for specific correlation information.

The processing identifier is information that identifies a process for aggregating questions. The processing identifier corresponds to, for example, processing information. The processing identifier may be the processing information itself. The processing identifier is, for example, "delete" or "integrate". The processing identifier "delete" is to delete a question. The processing identifier "integrate" is to integrate two or more questions. "Integrate" is to create a question that contains the contents of two or more questions from two or more questions. The process for aggregating questions is usually a process performed by the determination unit 133. The processing information is, for example, information that specifies a process for deleting a specific question. The processing information is, for example, information that specifies a process for integrating two or more specific questions.

The member score management unit 116 stores two or more member scores. It is preferable that the member score corresponds to a member identifier. It is preferable that the member score corresponds to an organization identifier. It is preferable that the member score corresponds to one or more organization attribute values for the organization identifier.

A member score is a score for each member. A member score has a total score. A member score usually has an answer score based on the answers to two or more questions. An answer score may be a score based on the answer to a question. An answer score may be the answer to a question itself. An answer score is, for example, an answer to a question asking about expectations or satisfaction with an item. An answer score is, for example, an item score. An item score is a score based on the answers to each question in a set of questions. An item score is, for example, a score based on the answer score to a question asking about expectations for an item and the answer score to a question asking about satisfaction with the item.

The total score constituting the member score of the member score management unit 116 may be information acquired by the total score acquisition means 1312, or may be information acquired by another device not shown and stored in the member score management unit 116. The answer score constituting the member score of the member score management unit 116 may be an answer accepted by the reception unit 12, or may be an answer score acquired by the processing unit 13 from an answer accepted by the reception unit 12, or may be an item score acquired by the answer score acquisition means 1311, or may be information acquired by another device not shown and stored in the member score management unit 116. In other words, the process by which the member scores are accumulated in the member score management unit 116 is not important.

The reception unit 12 receives various information and instructions. Examples of the various information and instructions include conditions, member scores, response information, questionnaire response information (member response information), and start instructions. Here, reception refers to, for example, reception from a terminal device (not shown), but may also be understood as a concept that includes reception of information input from an input device such as a keyboard, mouse, or touch panel, and reception of information read from a recording medium such as an optical disk, magnetic disk, or semiconductor memory.

The start instruction is an instruction to start processing by the question creation support device 1. It is preferable that the start instruction includes the conditions described below. Such conditions are conditions accepted by the condition acceptance unit 121.

The means for inputting various information and instructions can be anything, such as a touch panel, keyboard, mouse, or menu screen.

The condition receiving unit 121 receives conditions. The conditions are conditions for obtaining a member score. It is preferable that the conditions are organizational conditions. The organizational conditions are conditions using one or more organizational attribute values. For example, organizational conditions are "industry identifier = trading company", "size identifier = large company", and "department identifier 'Human resources OR Accounting'".

The means for inputting the conditions may be any means, such as a touch panel, keyboard, mouse, or menu screen. The condition receiving unit 121 may receive the organizational conditions from a terminal device (not shown).

The processing unit 13 performs various types of processing. The various types of processing are, for example, processing performed by the member score acquisition unit 131, the correlation acquisition unit 132, and the determination unit 133. The various types of processing are, for example, processing to accumulate received member scores, received response information, received questionnaire response information, received organizational information, etc. in the storage unit 11.

The member score acquisition unit 131 acquires a member score for each of the two or more members. Note that N is a natural number equal to or greater than three. The member score includes, for example, an answer score based on the member's answer to each of the N questions, and an overall score for the member. The member score acquisition unit 131 may also acquire an overall score for the organization.

The member score acquisition unit 131, for example, reads out one or more member scores stored in the member score management unit 116.

The member score acquisition unit 131, for example, acquires the member scores of one or more members who satisfy the conditions accepted by the condition acceptance unit 121. The member score acquisition unit 131, for example, acquires the member scores of two or more members who satisfy the conditions from the member score management unit 116.

It is preferable that each of the N questions described above corresponds to one classification identifier. The classification identifier is information that specifies the classification of the question. It is assumed that there are two or more classifications. In other words, each of the N questions corresponds to, for example, one of the two or more classification identifiers. Note that a classification may also be referred to as a class, group, category, factor, target, etc. Examples of classifications are "company foundation" and "philosophy strategy."

The answer score acquisition means 1311 acquires an answer score. For example, the answer score acquisition means 1311 acquires, for each question, an answer score based on the member's answer to the question. The answer score based on the answer may be the answer itself.

The answer score acquisition means 1311, for example, reads out the answer score held by the member score in the member score management unit 116.

The answer score acquisition means 1311 may acquire the answer score itself, or may read the answer score from the member score management unit 116.

The answer score acquisition means 1311, for example, for each organization, statistically processes the satisfaction level information contained in the answer information for each of two or more items contained in the answer information of the organization, and acquires an item score for each organization and each item. Note that the statistical processing may be, for example, a calculation of an average value, but is not essential. In such a case, the item score may be considered to be an answer score.

The answer score acquisition means 1311 acquires an item score for each organization, for example, using satisfaction information and expectation information for each item. Note that the items for which item scores are acquired using satisfaction information and expectation information are, for example, individual items. Note that satisfaction information is an answer to a question asking about satisfaction with an item. Expectation information is an answer to a question asking about expectations for an item.

It is preferable that the answer score acquisition means 1311 acquires item scores so that, for example, the higher the satisfaction information, the higher the score, and the lower the expectation information, the higher the score.

The answer score acquisition means 1311, for example, calculates the average value of the satisfaction information contained in the answer information and the average value of the expectation information contained in the answer information for each of two or more organizations and two or more items, and acquires an item score for each item using the two average values.

The answer score acquisition means 1311 may, for example, apply satisfaction information and expectation information to an individual score table to acquire an item score for each item. The answer score acquisition means 1311 may also, for example, apply a statistical processing result of satisfaction information and a statistical processing result of expectation information to an individual score table to acquire an item score for each item. The answer score acquisition means 1311 may, for example, apply an average value of satisfaction information and an average value of expectation information to an individual score table to acquire an item score for each item. The answer score acquisition means 1311 may, for example, apply a weighted average value of satisfaction information and a weighted average value of expectation information to an individual score table to acquire an item score for each item. The weighted average value is a weighted average value based on the attribute values of the members.

The overall score acquisition means 1312 acquires an overall score. For example, the overall score acquisition means 1312 reads out the overall score held by the member score of the member score management unit 116. The overall score acquisition means 1312 acquires an overall score, for example, using the item scores of two or more items acquired by the answer score acquisition means 1311. The overall score acquisition unit 132 typically acquires a higher overall score the better the scores of the two or more item scores.

The total score acquisition means 1312 acquires a total score for each of two or more members, for example, using the item scores for each of two or more items for that member. The total score acquisition means 1312 acquires a total score for the organization, for example, using the total scores for each of two or more members belonging to the organization.

The overall score acquisition means 1312 acquires one or two overall scores from the overall score of the member and the overall score of the organization. The overall score is acquired based on the answer scores. Here, the answer scores may be item scores.

The overall score acquisition unit 132 may calculate the overall score, for example, by "overall score = α × statistical score of item scores for overall items + β × statistical score of item scores for individual items." Here, it is preferable that (α>β). In other words, it is preferable that the overall score acquisition unit 132 compares the item scores for overall items with the item scores for individual items, and increases the weighting to acquire the overall score. Also, for example, "α=0.7, β=0.3." The statistical score of item scores for overall items is, for example, the average value or weighted average of item scores for overall items. The statistical score of item scores for individual items is, for example, the average value or weighted average of item scores for individual items.

The total score acquisition means 1312 may acquire the total score itself, or may read the total score from the member score management unit 116.

The correlation acquisition unit 132 acquires correlation information using two or more member scores acquired by the member score acquisition unit 131.

More specifically, the correlation acquisition unit 132 acquires correlation information regarding the correlation between items corresponding to one or more questions and the total score, for example, by using a member score that includes an answer score based on the member's answer to each of N questions for each of two or more members and a total score for the member. Note that N is a natural number equal to or greater than 3.

The correlation acquisition unit 132 acquires correlation information regarding the correlation between one or more questions and the total score, for example, by using a member score that includes an answer score based on the member's answer to each of N questions for each of two or more members and a total score for the member. Note that N is a natural number of 3 or more.

The correlation acquisition unit 132 acquires, for example, for each item, correlation information (which may be called partial correlation information) regarding the correlation between the answer score based on the answers to two or more questions for the item and the overall score, and acquires correlation information for the item based on the partial correlation information for each of the two or more questions. The larger the value of each of the two or more partial correlation information, the larger the correlation information for the item in such a case. The correlation information for the item is calculated, for example, by an increasing function that uses each of the two or more partial correlation information as a parameter.

For example, the correlation acquisition unit 132 acquires, for each item, first correlation information (which may be called first partial correlation information) regarding the correlation between the answer score based on the response regarding the expectation level for the item and the overall score, and acquires second correlation information (which may be called second partial correlation information) regarding the correlation between the answer score based on the response regarding the satisfaction level for the item and the overall score, and acquires correlation information for the item based on the first correlation information and the second correlation information.

The correlation acquisition unit 132 acquires the correlation information by, for example, any of the following methods (1), (2), or (3). That is, when acquiring the correlation information, the correlation acquisition unit 132 may use any algorithm such as a correlation function, a random forest, or a k-fold cross-validation.
(1) When using correlation functions (1-1) Basic processing when using correlation functions

The correlation acquisition unit 132 acquires, for example, the vectors of each of M members (answer score to question 1, answer score to question 2, ..., answer score to question N, total score), i.e. ( _I11 , _I12 , ..., _I1N , _O1 ), ( _I21 , _I22 , ..., _I2N , _O2 ), ... ( _IM1 , _IM2 , ..., _IMN , _OM ), from the member score management unit 116.

Next, the correlation acquisition unit 132 calculates a correlation coefficient for the total score of the answer scores for each question from question 1 to question N using a correlation function, and temporarily stores the correlation coefficient in a buffer (not shown) in association with each question identifier.
(1-2) Variation of (1-1)

The correlation acquisition unit 132 acquires, for example, for each of two or more items (group of questions) and for each of two or more questions corresponding to each item (e.g., a question asking about expectations, a question asking about satisfaction), the vectors (answer score to question 1, ..., answer score to question N, total score) of each of M members, ( _I11 , _I12 , ..., _I1N , _O1 ) ... ( _IM1 , _IM2 , ..., _IMN , _OM ) from the member score management unit 116.

Next, for each of two or more items (group of questions), and for each of two or more questions corresponding to each item, the correlation acquisition unit 132 calculates a correlation coefficient for the total score of the answer scores for each question from question 1 to question N using a correlation function, and temporarily stores the correlation coefficients in a buffer (not shown) in association with each question identifier.

Next, the correlation acquisition unit 132 acquires correlation information for each of two or more items (question groups) using an increasing function (for example, the sum) with the correlation coefficient for each of two or more questions corresponding to each item as a parameter, and temporarily stores the correlation coefficient in a buffer (not shown) in association with each item identifier. Note that, here, an item corresponds to a question group, which is a collection of two or more questions.
(2) When using Random Forest (2-1) Basic processing when using Random Forest

The correlation acquisition unit 132 acquires, for example, the vectors of each of M members (answer score to question 1, answer score to question 2, ..., answer score to question N, total score), i.e. ( _I11 , _I12 , ..., _I1N , _O1 ), ( _I21 , _I22 , ..., _I2N , _O2 ), ... ( _IM1 , _IM2 , ..., _IMN , _OM ), from the member score management unit 116.

Next, the correlation acquisition unit 132 constructs two or more pieces of teacher data for each vector of the members, with (answer score to question 1, answer score to question 2, ..., answer score to question N) as explanatory variables and the total score as the objective variable, and provides the two or more pieces of teacher data to a random forest learning module, executes the learning module, and constructs a learning model. Next, the correlation acquisition unit 132 acquires the influence of the answer score of each question from question 1 to question N on the total score in the learning model, and temporarily stores the influence in a buffer (not shown) in association with each question identifier. Note that the technology for acquiring the influence of each parameter in the random forest learning module is a publicly known technology.
(2-2) Variation of (2-1)

The correlation acquisition unit 132 acquires, for example, for each of two or more items (group of questions) and for each of two or more questions corresponding to each item (e.g., a question asking about expectations, a question asking about satisfaction), the vectors (answer score to question 1, ..., answer score to question N, total score) of each of M members, ( _I11 , _I12 , ..., _I1N , _O1 ) ... ( _IM1 , _IM2 , ..., _IMN , _OM ) from the member score management unit 116.

Next, the correlation acquisition unit 132 constructs two or more teacher data for each of two or more questions corresponding to each item, for each of two or more items (question groups), with (answer score of question 1, ..., answer score of question N) as explanatory variables and the total score as the objective variable, provides the two or more teacher data to a random forest learning module, executes the learning module, and constructs a learning model. Next, the correlation acquisition unit 132 acquires the influence of the answer score of each question from question 1 to question N on the total score in the learning model for each of two or more items (question groups), for each of two or more questions corresponding to each item, and temporarily stores the influence in a buffer (not shown) in association with each question identifier.

Next, the correlation acquisition unit 132 acquires correlation information for each of two or more items (question groups) using an increasing function (for example, the sum) with the degree of influence on each of two or more questions corresponding to each item as a parameter, and temporarily stores the correlation coefficient in a buffer (not shown) in association with each item identifier. Note that, here, an item corresponds to a question group, which is a collection of two or more questions.
(3) When using k-fold cross-validation (3-1) Basic processing when using k-fold cross-validation

The correlation acquisition unit 132 acquires, for example, the vectors of each of M members (answer score to question 1, answer score to question 2, ..., answer score to question N, total score), i.e. ( _I11 , _I12 , ..., _I1N , _O1 ), ( _I21 , _I22 , ..., _I2N , _O2 ), ... ( _IM1 , _IM2 , ..., _IMN , _OM ), from the member score management unit 116.

Next, the correlation acquisition unit 132 constructs teacher data for each of the N question identifiers, which is a vector of explanatory variables and objective variables excluding the answer score corresponding to that question identifier. Next, the correlation acquisition unit 132 acquires the accuracy (S _a ) of the learning model by k-fold cross-validation using the created teacher data for each of the N question identifiers. Next, the correlation acquisition unit 132 calculates correlation information for questions identified by each of the N question identifiers using a decreasing function (for example, "1/S _a ") with the accuracy (S _a ) corresponding to each question identifier as a parameter.
(3-2) Variation of (3-1)

The correlation acquisition unit 132 acquires, for example, for each of two or more items (group of questions) and for each of two or more questions corresponding to each item (e.g., a question asking about expectations, a question asking about satisfaction), the vectors (answer score to question 1, ..., answer score to question N, total score) of each of M members, ( _I11 , _I12 , ..., _I1N , _O1 ) ... ( _IM1 , _IM2 , ..., _IMN , _OM ) from the member score management unit 116.

Next, the correlation acquisition unit 132 configures teacher data, which is a vector of explanatory variables and objective variables excluding the answer scores corresponding to each of the two or more questions corresponding to each of the two or more items (question groups), for each of the two or more questions corresponding to each of the two or more items. Next, the correlation acquisition unit 132 acquires the accuracy (S _a ) of the learning model by k-fold cross-validation using the teacher data created for each of the two or more items (question groups) for each of the two or more questions corresponding to each of the two or more items. Next, the correlation acquisition unit 132 calculates the correlation information for each of the two or more items (question groups) by a decreasing function (e.g., "1/S _a ") with the accuracy (S _a ) corresponding to each of the two or more questions corresponding to each of the items as a parameter.

Next, the correlation acquisition unit 132 acquires correlation information for each of two or more items (question groups) using an increasing function (e.g., the sum) with the correlation information for each of two or more questions corresponding to each item as a parameter, and temporarily stores the correlation coefficients in a buffer (not shown) in association with each item identifier. Note that, in this case, an item corresponds to a question group, which is a collection of two or more questions.

The correlation acquisition unit 132 may acquire correlation information between two or more answer scores corresponding to different classification identifiers. That is, the correlation acquisition unit 132 acquires, for example, ( _I11 , I12, ..., I1N), (I21, I22, ..., _I2N ), ... (IM1, _IM2 , ..., IMN), which are vectors (answer score to question ₁ , answer score to question ₂ , ..., answer score to _{question N} ) of each of _{M members ,} from the member score management unit 116. Next, the distance between two of the vectors for each of the N questions (answer score for question 1 of member 1, answer score for question 1 of member 2, ..., answer score for question 1 of member M), (answer score for question 2 of member 1, answer score for question 2 of member 2, ..., answer score for question 2 of member M), ... (answer score for question N of member 1, answer score for question N of member 2, ..., answer score for question N of member M) is calculated, and two or more questions whose distance is within a threshold value are detected. Of these two or more questions, it is preferable that only one question is selected.

The decision unit 133 performs one or more of the deletion decision process or the integration decision process, and obtains decision information regarding the decision.

The deletion decision process is a process of deciding which of the N questions to delete. The decision unit 133 performs the deletion decision process of deciding which of the N questions to delete, for example, using correlation information for each of three or more item scores acquired by the correlation acquisition unit 132.

The integration decision process is a process of deciding two or more questions to be combined into one of the N questions. The decision unit 133 performs the integration decision process of deciding two or more questions to be combined into one of the N questions, for example, using correlation information for each of the three or more item scores acquired by the correlation acquisition unit 132.

The decision information is information acquired by the decision unit 133 as a result of processing. The decision information is, for example, information on questions to be deleted, information on questions to be merged, and questions after aggregation. The information on questions to be deleted is the question identifier of the question to be deleted, or the question to be deleted, or the question that remains without being deleted. The information on questions to be merged is the question identifier of each of two or more questions to be merged, or the question after integration. The questions after aggregation are the result of performing one or more of the deletion decision processing or integration decision processing on N questions, and are a set of questions with a reduced number. Aggregation means reducing the number of questions, for example, deletion or integration.

The decision unit 133, for example, uses the correlation information for each item to perform one or more of a deletion decision process that decides which of the N questions to delete, or a merge decision process that decides which of the N questions to merge into one, and acquires decision information.

For example, for each of two or more classifications, the determination unit 133 performs one or more of a deletion determination process that determines which questions to delete from among the M (3<=M<N) questions corresponding to the classification identifier that identifies the classification, or a merger determination process that determines which two or more questions to merge from among the M questions, and obtains the determination information.

The decision unit 133 performs one or more of the following processes: a deletion decision process that decides which questions to delete from M questions corresponding to the classification identifiers, so that the number of questions corresponding to each of the two or more classification identifiers is the same or the variation in the number of questions is below a threshold value; or an integration decision process that decides which two or more questions to merge into one of the M questions and acquires decision information regarding the decision. Note that the variation being below a threshold value means, for example, that the difference is below a threshold value (e.g., "1").

The decision unit 133, for example, for each of two or more classifications, obtains a correlation pattern identifier that identifies a pattern of correlation information of M answer scores corresponding to a classification identifier that identifies the classification, obtains a processing identifier that pairs with the correlation pattern identifier, performs a process identified by the processing identifier, and obtains decision information. The process identified by the processing identifier is, for example, either a deletion decision process or an integration decision process.

The decision unit 133, for example, uses correlation information between two or more answer scores to perform one or more of a deletion decision process that determines a question to be deleted that corresponds to at least one answer score among the two or more answer scores, or a merger decision process that determines two or more questions to be merged into one that corresponds to at least one item score among the two or more item scores, and acquires decision information.

The determination unit 133, for example, performs one or more of a deletion determination process for determining which of the X sets of question sets to delete, or a merge determination process for determining which of the X sets of question sets to merge, using correlation information between two or more answer scores, to obtain determination information. The question set has, for example, two questions asking about the expectation and satisfaction level for the item.

The determination unit 133, for example, uses the correlation information for each item to perform one or more of a deletion determination process for determining which question sets to delete from among the X question sets, or a merge determination process for determining which two or more question sets to merge from among the X question sets, and obtains the determination information.

The decision unit 133, for example, decides to delete one or more questions whose correlation information corresponds to an answer score that is equal to or less than a threshold. In such a case, the decision unit 133, for example, acquires an item identifier of the item corresponding to the question that has been decided to be deleted.

When the predetermined number of remaining items for each of two or more categories is R (e.g., 2), the decision unit 133 decides to delete questions corresponding to items corresponding to correlation information excluding the top R correlation information when, for example, correlation information other than the top R among multiple item scores corresponding to a category identifier satisfies a deletion condition. Note that the decision condition in such a case is, for example, "the difference between the Rth correlation information and the (R+1)th correlation information is equal to or greater than a threshold" and "the (R+1)th correlation information is equal to or less than a threshold".

When the predetermined number of remaining items for each of the two or more categories is R (e.g., 2), the decision unit 133 decides to merge two or more questions corresponding to the integration condition when, for example, multiple item scores corresponding to the category identifier satisfy the integration condition. Note that the integration condition is, for example, "the difference between the Rth correlation information and the (R+1)th correlation information is equal to or less than the threshold value (the difference is small)" or "all correlation information is equal to or greater than the threshold value."

The output unit 14 outputs the decision information acquired by the decision unit 133. Here, output is a concept that includes display on a display, projection using a projector, printing on a printer, sound output, transmission to an external device, storage on a recording medium, and delivery of the processing results to other processing devices or other programs, etc.

The storage unit 11, item information storage unit 111, answer information storage unit 112, individual score sheet storage unit 113, organizational information storage unit 114, pattern management unit 115, and member score management unit 116 are preferably non-volatile recording media, but can also be realized using volatile recording media.

The process by which information is stored in the storage unit 11, etc. is not important. For example, information may be stored in the storage unit 11, etc. via a recording medium, information transmitted via a communication line, etc. may be stored in the storage unit 11, etc., or information inputted via an input device may be stored in the storage unit 11, etc.

The reception unit 12 and the condition reception unit 121 may be realized by a device driver for an input means such as a touch panel or a keyboard, or control software for a menu screen. The reception unit 12 may also be realized by a wireless or wired communication means.

The processing unit 13, member score acquisition unit 131, correlation acquisition unit 132, determination unit 133, answer score acquisition means 1311, and total score acquisition means 1312 can usually be realized by a processor, memory, etc. The processing procedures of the processing unit 13, etc. are usually realized by software, and the software is recorded in a recording medium such as a ROM. However, they may also be realized by hardware (dedicated circuit). The processor may be a CPU, MPU, GPU, etc., and the type is not important.

The output unit 14 may or may not include an output device such as a display or speaker. The output unit 14 may be realized by driver software for an output device, or by a combination of driver software for an output device and an output device, etc.

Next, an example of the operation of the question creation support device 1 will be explained using the flowchart in Figure 2.

(Step S201) The reception unit 12 determines whether or not a start instruction has been received. If a start instruction has been received, the process proceeds to step S202; if a start instruction has not been received, the process returns to step S20.

(Step S202) The member score acquisition unit 131 determines whether the start instruction received in step S201 includes a condition. If a condition is included, the process proceeds to step S203; if a condition is not included, the process proceeds to step S204.

(Step S203) The member score acquisition unit 131 acquires the conditions included in the start instruction received in step S201.

(Step S204) The member score acquisition unit 131 acquires two or more member scores from the member score management unit 116. When moving from step S203 to step S204, the member score acquisition unit 131 acquires two or more member scores that meet the conditions. When moving from step S202 to step S204, the member score acquisition unit 131 acquires all member scores from the member score management unit 116.

(Step S205) The correlation acquisition unit 132 and the determination unit 133 perform a determination process using the two or more member scores acquired in step S204, and acquire determination information. An example of the determination process will be described with reference to the flowcharts in Figures 3 and 6.

(Step S206) The output unit 14 outputs the decision information acquired in step S205. Return to step S201.

In the flowchart in Figure 2, processing ends when the power is turned off or an interrupt occurs to end processing.

Next, a first example of the determination process in step S205 will be described using the flowchart in FIG. 3. The determination process here is an example of a case where multiple classifications typically exist.

(Step S301) The correlation acquisition unit 132 assigns 1 to counter i.

(Step S302) The correlation acquisition unit 132 determines whether the i-th classification exists. If the i-th classification exists, the process proceeds to step S303. If the i-th classification does not exist, the process returns to the upper level process.

(Step S303) The correlation acquisition unit 132 acquires, from the member scores acquired in step S204, a member score that has an answer score and a total score that are paired with a classification identifier that identifies the i-th classification.

(Step S304) The correlation acquisition unit 132 assigns 1 to counter j.

(Step S305) The correlation acquisition unit 132 determines whether or not the jth item exists in the ith classification. If the jth item exists, the process proceeds to step S306, and if the jth item does not exist, the process proceeds to step S312. Note that an item usually corresponds to two or more questions, but may correspond to one question.

(Step S306) The correlation acquisition unit 132 assigns 1 to the counter k.

(Step S307) The correlation acquisition unit 132 determines whether the kth question exists in the jth item. If the kth question exists, the process proceeds to step S308; if the kth question does not exist, the process proceeds to step S310.

(Step S308) The correlation acquisition unit 132 acquires correlation information for the total score of the answer score to the kth question in the jth item.

(Step S309) The correlation acquisition unit 132 increments the counter k by 1. Return to step S307.

(Step S310) The correlation acquisition unit 132 uses the one or more pieces of correlation information acquired in step S308 to acquire correlation information for the jth item, the correlation information being larger the larger each piece of correlation information is.

If two or more pieces of correlation information are acquired in step S308, the correlation acquisition unit 132 calculates the correlation information for the j-th item, for example, by an increasing function with the two or more pieces of correlation information as parameters. If only one piece of correlation information is acquired in step S308, the correlation acquisition unit 132 acquires the one piece of correlation information as the correlation information for the j-th item.

(Step S311) The correlation acquisition unit 132 increments the counter j by 1. Return to step S305.

(Step S312) The determination unit 133 performs a classification aggregation process, which is a process of aggregating items corresponding to the i-th classification, using the correlation information for each item acquired in step S310. An example of such a classification aggregation process will be described with reference to the flowchart in FIG. 4. Note that the classification aggregation process is a process for aggregating questions corresponding to one classification. Aggregating questions may also mean aggregating items.

(Step S313) The decision unit 133 uses the results of the classification aggregation process in step S312 to obtain decision information corresponding to the i-th classification. An example of such decision information acquisition process is described using the flowchart in FIG. 5.

(Step S314) The correlation acquisition unit 132 increments the counter i by 1. Return to step S302.

Next, an example of the classification aggregation process in step S312 will be described using the flowchart in FIG.

(Step S401) The determination unit 133 acquires a correlation information vector, which is a vector whose elements are the correlation information for each item acquired in step S310. Note that the correlation information vector is, for example, a vector whose elements are the correlation information for each item of the target classification, for example (correlation information for item 1, correlation information for item 2, ..., correlation information for item X). Also, X is, for example, "4".

(Step S402) The determination unit 133 assigns 1 to the counter i.

(Step S403) The determination unit 133 determines whether the i-th pattern information exists in the pattern management unit 115. If the i-th pattern information exists, the process proceeds to step S404; if not, the process proceeds to step S408.

(Step S404) The determination unit 133 obtains the correlation pattern information contained in the i-th pattern information from the pattern management unit 115.

(Step S405) The determination unit 133 determines whether the correlation information vector acquired in step S401 matches the pattern indicated by the correlation pattern information acquired in step S404. If it matches the pattern, the process proceeds to step S406, and if it does not match, the process proceeds to step S407.

(Step S406) The determination unit 133 obtains a processing identifier that pairs with the correlation pattern information obtained in step S404 from the pattern management unit 115.

(Step S407) The determination unit 133 increments the counter i by 1. Return to step S403.

(Step S408) The determination unit 133 obtains the default process identifier from the pattern management unit 115. Then, the process returns to the upper level process.

Next, an example of the decision information acquisition process in step S313 will be described using the flowchart in FIG.

(Step S501) The determination unit 133 obtains multiple questions that are paired with the target classification identifier.

(Step S502) The determination unit 133 obtains processing information identified by the processing identifier obtained in step S406 from the pattern management unit 115.

(Step S503) The determination unit 133 determines whether the processing information acquired in step S502 includes a deletion process (e.g., "delete"). If it includes a deletion process, the process proceeds to step S504; if it does not, the process proceeds to step S506.

(Step S504) The determination unit 133 acquires item identifiers of one or more items to be deleted, which are item identifiers specified by the deletion process included in the processing information acquired in step S502. Note that the item identifiers may be question identifiers.

(Step S505) The determination unit 133 deletes the questions that correspond to one or more item identifiers obtained in step S504 from the multiple questions obtained in step S501.

(Step S506) The determination unit 133 assigns 1 to the counter i.

(Step S507) The determination unit 133 determines whether the processing information acquired in step S502 includes the i-th integration process (e.g., "integrate"). If it does include the i-th integration process, the process proceeds to step S508; if it does not, the process returns to the upper process.

(Step S508) The determination unit 133 obtains the item identifiers of the two or more items to be integrated, which are two or more item identifiers specified by the i-th integration process.

(Step S509) The determination unit 133 obtains questions corresponding to the two or more item identifiers obtained in step S508.

(Step S510) The determination unit 133 merges the two or more questions obtained in step S509.

(Step S511) The determination unit 133 increments the counter i by 1. Return to step S507.

In the flowchart of FIG. 5, one or more aggregated questions corresponding to the field identified by the target classification identifier are obtained.

Next, a second example of the determination process in step S205 will be described using the flowchart in FIG. 6. The determination process here is an example of a case where no classification exists.

(Step S601) The determination unit 133 assigns 1 to the counter i.

(Step S602) The determination unit 133 determines whether the i-th item exists. If the i-th item exists, the process proceeds to step S603. If the i-th item does not exist, the process proceeds to step S605. Note that the item may be a question.

(Step S603) The determination unit 133 obtains correlation information for the answer score for the i-th item with the integrated score using the answer score and the integrated score for the i-th item held by each of the multiple member scores obtained in step S204.

(Step S604) The determination unit 133 increments the counter i by 1. Return to step S602.

(Step S605) The determination unit 133 performs a process of aggregating the questions (items in this case) using the correlation information for each of the two or more items acquired in step S603. The process returns to the upper level process. An example of such aggregation process will be described with reference to the flowchart in FIG. 7.

Next, an example of the aggregation process in step S605 will be described using the flowchart in FIG.

(Step S701) The determination unit 133 sorts the questions in the item information storage unit 111 in descending order using the correlation information as a key.

(Step S702) The determination unit 133 deletes questions that are paired with correlation information whose correlation information is equal to or less than the threshold value α from the set of questions.

(Step S703) The determination unit 133 acquires questions from the set of questions that are paired with correlation information whose correlation information is equal to or greater than the threshold value β. Note that "β>α".

(Step S704) The determination unit 133 determines whether to create other questions in addition to the questions adopted in step S703. If other questions are to be created, the process proceeds to step S705; if not, the process returns to the upper level process.

Whether or not to create another question depends on, for example, the number of questions used being below the minimum number of questions, or if there is a question whose correlation information is equal to or greater than the threshold γ or is paired with correlation information greater than the threshold γ. Note that γ<β.

(Step S705) The determination unit 133 selects the two questions with the highest correlation information from among the unadopted questions, and creates a single question by combining the two questions.

The following describes a specific example of the operation of the question creation support device 1 in this embodiment.

The item information storage unit 111 of the question creation support device 1 stores the item information management table shown in FIG. 8. The item information management table is a table that manages a large number of item information items that indicate the items of a questionnaire for members (employees in this case) of an organization (a company in this case). It can be said that the item information management table is information that indicates the questions of the questionnaire for members. Here, the item information has "item identifier", "type", "classification", "item", "question: expectation level", and "question: satisfaction level". The "item identifier" is an ID that identifies the item. The "type" is information that indicates the type of item, and here, it can be either a comprehensive item or an individual item. The "classification" is an intermediate concept of the item, and may be called an object, factor, etc. The "item" is information that indicates the content of the item. It is also possible to consider the "item" as an item identifier. The "question: expectation level" is a question for acquiring expectation level information. The "question: satisfaction level" is a question for acquiring satisfaction level information. In other words, here, for one item, there is a question group that is a set of two questions. The set of questions includes questions asking about expectations for one item and questions asking about satisfaction with one item.

The item information management table (Figure 8) has four overall items and 64 individual items. The 64 individual items are divided into 16 categories. Each of the 16 categories contains four items. Each item corresponds to a question asking about the expectation for the item and a question asking about the satisfaction level for the item.

The answer information storage unit 112 stores answer information having a structure as shown in FIG. 9, for example. The answer information storage unit 112 stores two or more pieces of answer information. FIG. 9 shows the results of a survey of many members of an organization. In FIG. 9, one table is a collection of answer information for one member. That is, 701 is the answer information for one member. The answer information has many records (here, 63 or more) having "item identifier", "expectation information", and "satisfaction information". The records of items with item identifiers 1 to 4 are records of comprehensive items and do not have expectation information. The expectation information and satisfaction information constituting the answer information of 701 are information obtained from the answers given by the members to "Question: expectation" and "Question: satisfaction" in the item information management table shown in FIG. 6. Such answers are answers in the form of any of the natural numbers from 1 to 5. Here, when the expectation information for "question: expectation" is 1, the expectation is the lowest, and when it is 5, the expectation is the highest. In addition, when the satisfaction level information for "Question: Satisfaction" is 1, the satisfaction level is the lowest, and when it is 5, the satisfaction level is the highest. Each of the expectation level information and expectation level information is an example of an answer score.

The individual score table storage unit 113 stores the individual score table shown in FIG. 10. The individual score table manages two or more records having "expectation information", "satisfaction information", and "item score". The attribute values of "expectation information" are "expectation value 1", "expectation value 2", ... "expectation value N", which are specific values or range information. The attribute values of "satisfaction information" are "satisfaction value 1", "satisfaction value 2", ... "satisfaction value N", which are specific values or range information. "Item score" here is information indicating the item score. The attribute values of "item score" are "score 1", "score 2", ... "score N", which are specific values. In other words, the individual score table is a table for determining item scores based on expectation information and satisfaction information for an item. When the expectation of a member's answer to a certain item is "expectation value 1" and the satisfaction level is "satisfaction value 1", the answer score acquisition means 1311 acquires, for example, score 1, which is paired with "expectation value 1" and "satisfaction value 1", as the item score of the item.

The pattern management unit 115 stores a pattern management table shown in FIG. 11. The pattern management table is a table for managing pattern information. The pattern management table manages two or more records having a "correlation pattern identifier", a "correlation pattern name", "correlation pattern information", a "processing identifier", and "processing information". The "correlation pattern identifier" is the ID of the correlation pattern. The "correlation pattern name" is the name of the correlation pattern. "C (2nd place)" in the "correlation pattern information" indicates that the correlation information is ranked second among the four items in the classification. The correlation pattern information "C (2nd place) - C (3rd place) >= T ₁ " for "ID=1" indicates that the difference between the correlation information ranked second and the correlation information ranked third is T ₁ or more (the top two correlation information are significantly higher than the others). The correlation pattern information of "ID=2""C (2nd place) - C (3rd place) <= T ₂ " indicates that the difference between the correlation information of the 2nd place and the correlation information of the 3rd place is T ₂ or less (the difference between the correlation information of the 2nd place and the correlation information of the 3rd place is considerably small). The correlation pattern information of "ID=3""C (4th place) >= T ₃ " indicates that the correlation information of the 4th place is T ₃ or more (all correlation information is considerably high). The correlation pattern information of "ID=4""C (1st place) >= T ₄ AND C (2nd place) - C (4th place) <= T ₅ " indicates that the correlation information of the 1st place is T ₄ or more, and the difference between the correlation information of the 2nd place and the 4th place is T ₅ or less (the 1st place is considerably high, and the 2nd place and below are low). The correlation pattern information of "ID=5""-" indicates that it is default correlation pattern information that is applied when there is no match with any of the correlation pattern information of "ID=1 to 4". The "processing identifier" is the ID of the process to be performed. The "processing information" indicates the content of the processing to be performed. The "delete" included in the processing information indicates a deletion process. The "Integrate" indicates an integration process. For example, the "delete" (third place) indicates that one or more questions corresponding to the item corresponding to the third-place correlation information are deleted. For example, the "Integrate" (second place, third place) indicates that an item corresponding to the second-place correlation information and an item corresponding to the third-place correlation information are integrated to create a new item. Note that creating a new item here means creating a question group consisting of two questions, one asking about expectation and one asking about satisfaction. Creating a new item means creating a new question. The processing information of "ID=1" indicates that the correlation information leaves the top two questions. The processing information of "ID=2" indicates that the correlation information leaves the question in the first place, integrates the questions in the second place and the third place, and deletes the question in the fourth place. The processing information of "ID=3" indicates that the correlation information integrates the questions in the first place and the second place, and integrates the questions in the third place and the fourth place. The processing information of "ID=4" indicates that the correlation information keeps the question in first place and combines the three questions in second, third, and fourth places. The processing information of "ID=5" indicates that the correlation information combines the questions in first and fourth places, and combines the questions in second and third places.

In this specific example, C (xth place) in Fig. 11 is the xth place correlation information among the correlation information of items calculated by the sum of two answer scores to each question of expectation and satisfaction. In this specific example, the threshold _T1 in Fig. 11 is "0.3", the threshold _T2 is "0.1", and the threshold _T3 is "1.4".

It is assumed that the member score management table shown in FIG. 12 is stored in the member score management unit 116. The member score management table stores a large number of records having an "ID," "member identifier," "answer score," and "total score." In this case, the "answer score" is the expectation and satisfaction level for each of the individual items 64 (items 5 to 68). It is assumed that the member scores in the member score management table are acquired, for example, by the answer score acquisition means 1311 and the total score acquisition means 1312, and stored in the member score management table.

In the above situation, it is assumed that the reception unit 12 of the question creation support device 1 has received a start instruction.

Then, the member score acquisition unit 131 acquires all records (member scores) from the member score management table (Figure 12).

Next, the correlation acquisition unit 132 uses the many member scores acquired by the member score acquisition unit 131 to acquire correlation information between the answer score (i.e., expectation or satisfaction) and the overall score for each classification identifier and for each question. Note that here, the correlation information is, for example, a correlation coefficient. Also, here, the correlation coefficient is referred to as the degree of correlation.

Now, for example, suppose that questions in a category corresponding to the category identifier "ideal strategy" are to be aggregated. The items corresponding to the category identifier "ideal strategy" are shown in Figure 13. In Figure 13, the category identified by the category identifier "ideal strategy" has four items with item identifiers "5" to "8". Each item has a question asking about expectations and a question asking about satisfaction, so there are eight questions corresponding to that category.

For example, the correlation acquisition unit 132 acquires a correlation between the answer information "expectations" of the item identifier "5" paired with the classification identifier "ideal strategy" and the overall score (here, engagement score) of 0.6, and a correlation between the answer information "satisfaction" of the item identifier "5" and the overall score of 0.7. Also, it is assumed that the correlation between the answer information "expectations" of the item identifier "6" paired with the classification identifier "ideal strategy" and the overall score of 0.7, and a correlation between the answer information "satisfaction" of the item identifier "6" and the overall score of 0.7. Also, it is assumed that the correlation between the answer information "expectations" of the item identifier "7" paired with the classification identifier "ideal strategy" and the overall score of 0.9, and a correlation between the answer information "satisfaction" of the item identifier "7" and the overall score of 0.9. Furthermore, suppose that the correlation between the answer information "Expectations" of the item identifier "8" that is paired with the classification identifier "Philosophy Strategy" and the overall score is 0.5, and the correlation between the answer information "Satisfaction" of the item identifier "8" and the overall score is 0.5.

Next, the correlation acquisition unit 132 acquires correlation information corresponding to each item, for example, by using the correlation between the two answer scores (expectations, satisfaction) corresponding to each item. Note that here, the correlation acquisition unit 132 regards the sum of the two correlations as the correlation information corresponding to the item. In other words, the correlation acquisition unit 132 acquires correlation information "1.3" for the item with item identifier "5", correlation information "1.4" for the item with item identifier "6", correlation information "1.8" for the item with item identifier "7", and correlation information "1.0" for the item with item identifier "5". Note that such correlation information is the "correlation information" in FIG. 13.

Next, the determination unit 133 obtains a correlation information vector (1.3, 1.4, 1.8, 1.0), which is a vector whose elements are correlation information for the obtained item "ideal strategy."

Next, the determination unit 133 refers to the pattern management table and determines that the correlation information vector (1.3, 1.4, 1.8, 1.0) matches the second pattern information. Next, the determination unit 133 obtains the processing identifier "PR2" from the pattern management table.

Next, the determination unit 133 obtains the process information "Integrate (2nd place, 3rd place) delete (4th place)" that is paired with the process identifier "PR2" from the pattern management table.

Next, the determination unit 133 integrates the items with the second and third correlation information items according to the processing information "Integrate (2nd, 3rd)." In other words, the determination unit 133 obtains a new item, "The company makes an effort to instill its corporate philosophy and vision in employees, and instills its corporate philosophy and vision in employees," according to "Integrate (2nd, 3rd)."

The determination unit 133 also deletes the item with the correlation information in fourth place according to the processing information "delete (fourth place)." In other words, the determination unit 133 deletes the item identified by the item identifier "8" according to "delete (fourth place)."

As a result of the above processing, the number of items for the category "Philosophy Strategy" was reduced from four to two. The content of the items was: (A) "The company makes an effort to instill its corporate philosophy and vision in employees, and instills its corporate philosophy and vision in employees," and (B) "The company tries to communicate its current strategies and goals to employees in an easy-to-understand manner." In addition, there were four questions: one asking about expectations for item (A), one asking about satisfaction with item (A), one asking about expectations for item (B), and one asking about satisfaction with item (B).

The correlation acquisition unit 132 and the determination unit 133 perform the above process on the other classifications, and the number of items in the other classifications is also consolidated from 4 to 2. As a result, the number of items in the 64 individual items is reduced to 32. In other words, the number of questions corresponding to the individual items is reduced from 128 to 64.

As described above, according to this embodiment, an appropriate number of appropriate questions can be proposed, thereby reducing the burden on members when obtaining a total score.

In addition, this embodiment ensures a balance in the number of questions in each category, resulting in a well-balanced and appropriate set of questions when obtaining an overall score.

Furthermore, according to this embodiment, it is possible to propose an appropriate set of questions for each condition. It is preferable that the conditions are organizational conditions based on one or more organizational attribute values.

In addition, this embodiment reduces the burden on members when obtaining a total score based on a questionnaire asking about their expectations and satisfaction with items.

In this embodiment, the question creation support device 1 may be a server accessible from a terminal device (not shown). In such a case, the reception unit 12 receives various instructions and information from the terminal device. The output unit 14 transmits various information to the terminal device.

In the present embodiment, the question creation support device 1 may be a device that accesses the member score management unit 116 present in an external device, and the correlation acquisition unit 132, the decision unit 133, and the output unit 14 perform the above-described processing. In such a case, the question creation support device 1 is a device that includes a correlation acquisition unit 132 that acquires correlation information regarding the correlation between items corresponding to one or more questions and a total score using a member score having an answer score based on the member's answer to each of N questions (N is a natural number of 3 or more) for each of two or more members and a total score of the member, a decision unit 133 that performs one or more processes of a deletion decision process that determines a question to be deleted from the N questions, or an integration decision process that determines two or more questions to be integrated into one of the N questions, using the correlation information for each item, and acquires decision information regarding the decision, and an output unit 14 that outputs the decision information.

Furthermore, the processing in this embodiment may be realized by software. This software may be distributed by software download or the like. This software may also be recorded on a recording medium such as a CD-ROM and distributed. This also applies to other embodiments in this specification. The software that realizes the question creation support device 1 in this embodiment is the following program. In other words, this program causes a computer to function as a correlation acquisition unit that acquires correlation information regarding the correlation between items corresponding to one or more questions and the overall score using a member score that has an answer score based on the member's answer to each of N (N is a natural number of 3 or more) questions for each of two or more members and a total score for the members, a decision unit that performs one or more processes of a deletion decision process that determines a question to be deleted from the N questions using the correlation information for each item, or an integration decision process that determines two or more questions to be integrated into one of the N questions, acquires decision information regarding the decision, and an output unit that outputs the decision information.

FIG. 14 shows the appearance of a computer that executes the program described in this specification to realize the question creation support device 1 and the like of the various embodiments described above. The above-mentioned embodiments can be realized by computer hardware and a computer program executed thereon. FIG. 14 is an overview of this computer system 300, and FIG. 15 is a block diagram of system 300.

In FIG. 14, computer system 300 includes computer 301, which includes a CD-ROM drive, keyboard 302, mouse 303, and monitor 304.

In FIG. 15, in addition to a CD-ROM drive 3012, computer 301 includes an MPU 3013, a bus 3014 connected to the CD-ROM drive 3012 etc., a ROM 3015 for storing programs such as a boot-up program, a RAM 3016 connected to the MPU 3013 for temporarily storing instructions for application programs and providing temporary storage space, and a hard disk 3017 for storing application programs, system programs, and data. Although not shown here, computer 301 may further include a network card that provides connection to a LAN.

A program that causes computer system 300 to execute functions such as the question creation support device 1 of the above-mentioned embodiment may be stored on CD-ROM 3101, inserted into CD-ROM drive 3012, and then transferred to hard disk 3017. Alternatively, the program may be sent to computer 301 via a network (not shown) and stored on hard disk 3017. The program is loaded into RAM 3016 when executed. The program may be loaded directly from CD-ROM 3101 or the network.

The program does not necessarily have to include an operating system (OS) that causes the computer 301 to execute the functions of the question creation support device 1 of the above-mentioned embodiment, or a third-party program. The program only needs to include an instruction portion that calls appropriate functions (modules) in a controlled manner to obtain the desired results. How the computer system 300 operates is well known, and a detailed description will be omitted.

The program may be executed by a single computer or multiple computers. In other words, the program may be executed by centralized processing or distributed processing.

Furthermore, in each of the above embodiments, it goes without saying that two or more communication means present in one device may be realized physically by one medium.

In addition, in each of the above embodiments, each process may be realized by centralized processing in a single device, or may be realized by distributed processing in multiple devices.

The present invention is not limited to the above-described embodiment, and various modifications are possible, and it goes without saying that these are also included within the scope of the present invention.

As described above, the question creation support device 1 of the present invention has the effect of being able to propose an appropriate number of questions, thereby reducing the burden on members when obtaining a total score, and is useful as a device to support the aggregation of questions, etc.

REFERENCE SIGNS LIST 1 Question creation support device 11 Storage unit 12 Reception unit 13 Processing unit
14 Output unit 111 Item information storage unit 112 Answer information storage unit 113 Individual score table storage unit 114 Organization information storage unit 115 Pattern management unit 116 Member score management unit 121 Condition reception unit 131 Member score acquisition unit 132 Correlation acquisition unit 132 Total score acquisition unit 133 Determination unit 1311 Answer score acquisition means 1312 Total score acquisition means

Claims (12)

a correlation acquisition unit that acquires correlation information regarding the correlation between an item corresponding to one or more questions and a total score using a member score having an answer score based on the member's answer to each of N questions (N is a natural number equal to or greater than 3) for each of two or more members and a total score for the member;
a determination unit that performs one or more of a deletion determination process for determining a question to be deleted from the N number of questions, or a merger determination process for determining two or more questions to be merged into one of the N number of questions, using the correlation information for each of the items, and acquires decision information regarding the decision;
and an output unit that outputs the determination information. The method further includes a member score acquisition unit that acquires, for each of the two or more members, a member score having an answer score based on the member's answer to each of the N questions and a total score for the member,
2. The question creation support device according to claim 1, wherein the two or more member scores used by the correlation acquisition unit are two or more member scores acquired by the member score acquisition unit. Each of the N questions is associated with a classification identifier that identifies one of two or more classifications of the question;
The determination unit is
A question creation support device as described in claim 1 or claim 2, wherein for each of two or more classifications, one or more of a deletion decision process that decides which questions to delete from among M (3 <= M < N) questions corresponding to a classification identifier that identifies the classification, or a merger decision process that decides which two or more questions to merge from among the M questions, is performed to obtain the decision information. The determination unit is
The question creation support device of claim 3 performs one or more of the following processes: a deletion decision process that determines which questions to delete from among the M questions corresponding to the classification identifiers so that the number of questions corresponding to each of the two or more classification identifiers is the same or the variation is below a threshold value, or an integration decision process that determines which two or more questions to merge into one of the M questions and obtains decision information regarding the decision. The answer score corresponding to at least one question corresponds to any one of the classification identifiers;
a process identifier that specifies a process to be performed by the determination unit is associated with each of two or more correlation pattern identifiers that specify a pattern of correlation information of the M answer scores corresponding to the same classification identifier;
the process specified by the process identifier is either the deletion decision process or the integration decision process,
The determination unit is
The question creation support device of claim 3, further comprising: a correlation pattern identifier for identifying a pattern of correlation information of M answer scores corresponding to a classification identifier for identifying the classification, a processing identifier that pairs with the correlation pattern identifier, a processing specified by the processing identifier, and the determination information. the member score corresponds to one or more organizational attribute values;
The member score acquisition unit,
Obtaining a member score for each of two or more members who satisfy an organizational condition using one or more organizational attribute values;
The correlation acquisition unit
The question creation support device according to claim 2 , wherein the correlation information is acquired by using the two or more member scores acquired by the member score acquisition unit. Each of the N questions corresponds to one of two or more classification identifiers;
Each of the three or more answer scores corresponding to at least one question corresponds to one of the classification identifiers;
The correlation acquisition unit
Obtaining correlation information between two or more answer scores corresponding to different classification identifiers;
The determination unit is
A question creation support device as described in claim 1 or claim 2, wherein correlation information between the two or more answer scores is used to perform one or more of a deletion decision process that determines a question to be deleted, which is a question corresponding to at least one of the two or more answer scores, or a merger decision process that determines two or more questions to be merged into one, which are two or more questions corresponding to at least one of the two or more answer scores, and acquires the decision information. The N questions are a set of X (X>=3) questions each having two or more questions and corresponding to an item;
The determination unit is
2. The question creation support device according to claim 1, further comprising: a deletion decision process for deciding which question sets to delete from the X sets of question sets; or an integration decision process for deciding two or more question sets to integrate from the X sets of question sets, using the correlation information of the two or more answer scores, to acquire the decision information. The set of questions includes two questions asking about expectation and satisfaction with an item,
9. The question creation support device according to claim 8, wherein the total score is obtained using item scores obtained using two answers, the answer to the expectation for the item and the answer to the satisfaction for the item. The correlation acquisition unit
obtain, for each of the items, first correlation information regarding a correlation between an answer score based on the answer of the expectation for the item and a total score, and obtain second correlation information regarding a correlation between an answer score based on the answer of the satisfaction for the item and a total score, and obtain correlation information based on the first correlation information and the second correlation information;
The determination unit is
The question creation support device according to claim 9, further comprising: a deletion decision process for deciding which of the X sets of question sets to delete, or an integration decision process for deciding two or more of the X sets of question sets to integrate, using the correlation information for each of the items, to acquire the decision information. A question creation support method realized by a correlation acquisition unit, a determination unit, and an output unit,
a correlation acquisition step in which the correlation acquisition unit acquires correlation information regarding the correlation between an item corresponding to one or more questions and a total score by using a member score having an answer score based on the member's answer to each of N questions (N is a natural number equal to or greater than 3) for each of two or more members and a total score for the member;
a determination step in which the determination unit performs one or more of a deletion determination process for determining a question to be deleted from the N number of questions, or a merger determination process for determining two or more questions to be merged into one of the N number of questions, using the correlation information for each of the items, and acquires determination information regarding the determination;
and an output step in which the output unit outputs the decision information. Computer,
a correlation acquisition unit that acquires correlation information regarding the correlation between an item corresponding to one or more questions and a total score using a member score having an answer score based on the member's answer to each of N questions (N is a natural number equal to or greater than 3) for each of two or more members and a total score for the member;
a determination unit that performs one or more of a deletion determination process for determining a question to be deleted from the N number of questions, or a merger determination process for determining two or more questions to be merged into one of the N number of questions, using the correlation information for each of the items, and acquires decision information regarding the decision;
A program for causing the device to function as an output unit that outputs the decision information.

Images