JP5271821B2 - Investigation device and computer program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To contribute to acquisition of a change in the attitude of an examinee group on a survey target caused by a stimulus. <P>SOLUTION: A survey device includes: a survey execution module 2 for repetitively executing a survey stage to obtain evaluation on the survey target by presenting information related to the survey target, while presenting the stimulus based on the evaluation obtained in the survey stage to the examinee at a new survey stage; a raw data storage unit 7 for storing information related to the obtained evaluation; a process data storage unit 8; and a survey result analysis module 5 for generating analysis data related to the change of evaluation between the survey stages. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to an investigation apparatus and a computer program.

  In general, consumer evaluation and purchase behavior for products and services change in response to promotional activities conducted by companies that provide products and information from consumer generated media (CGM) such as word-of-mouth spread on the Internet. There are many. In particular, since the role played by CGM has increased in recent years, it is desirable to consider the impact of CGM when investigating consumer evaluation of goods and services. Conventionally, for example, an investigation apparatus described in Patent Document 1 is known. In this conventional research apparatus, the existing evaluation for the survey target is presented to the mutual evaluation subject, the evaluation for the existing evaluation is obtained, and the existing evaluation and the evaluation for the existing evaluation are recorded in association with each other. Yes.

JP 2009-59206 A

  However, in the conventional survey apparatus described above, the evaluation of the subject with respect to the survey target is acquired before and after the presentation of the existing evaluation, but only this grasps how the attitude of the subject group with respect to the survey target changes. Is difficult. For this reason, it is difficult to fully analyze the impact of CGM on consumers.

  The present invention has been made in consideration of such circumstances, and the purpose thereof is an investigation device that can contribute to grasping how the attitude of the subject group with respect to the investigation object changes due to the stimulus, and To provide a computer program.

  In order to solve the above-mentioned problem, the investigation device according to the present invention provides the evaluation stage acquired at the investigation stage, which presents information on the investigation object to the subject group and obtains the evaluation on the investigation object. While presenting the stimuli based on the novel to the subject in the investigation stage, the storage means for repeatedly storing, the storage means for storing the information on the acquired evaluation, and the analysis data on the change in the evaluation between the investigation stages Analysis data generating means for generating.

The investigation apparatus according to the present invention is characterized by comprising comprehensive evaluation aggregation means for total evaluation for each investigation stage.
The investigation apparatus according to the present invention is characterized by comprising partial evaluation aggregation means for totalizing partial evaluations for each of the investigation stages.
The investigation apparatus according to the present invention is characterized by comprising partial evaluation importance calculation means for calculating the importance of partial evaluation for each of the investigation stages.

The investigation apparatus according to the present invention is characterized by comprising partial evaluation aggregation means for totalizing partial evaluations for each of the investigation stages and for each comprehensive evaluation.
The investigation apparatus according to the present invention is characterized by comprising partial evaluation importance calculation means for calculating importance of partial evaluation for each investigation stage and for each comprehensive evaluation.
In the investigation apparatus according to the present invention, it is provided with a stimulus totaling unit that aggregates the stimuli presented to the subject by distinguishing the subject who has changed the overall evaluation between the investigation stages and the subject who has not changed. And

In the investigation apparatus according to the present invention, there is provided a stimulus importance degree calculation means for distinguishing between a subject who has changed in the overall evaluation and a subject who has not changed between the investigation stages and calculating the importance degree of the stimulus presented to the subject. It is characterized by having.
In the investigation device according to the present invention, the stimulus totaling means totals stimuli presented to a subject whose overall evaluation has changed between the investigation stages according to the direction of change.
In the investigation device according to the present invention, the stimulus importance calculation means calculates the importance of the stimulus presented to the subject whose overall evaluation has changed between the investigation stages, according to the direction of change. .

In the investigation apparatus according to the present invention, the subject attribute of the subject who performed the evaluation based on the stimulus presented to the subject by distinguishing the subject who had a change in the overall evaluation from the subject who did not change between the investigation stages. It is characterized by having an attribute totaling means for totaling.
The investigation apparatus according to the present invention is characterized by comprising attribute counting means for distinguishing between subjects whose overall evaluation has changed between the investigation stages and subjects who have not changed, and totaling the attributes of subjects.
In the investigation device according to the present invention, the attribute counting means totals the attributes for each direction of change.

The investigation apparatus according to the present invention is characterized by comprising attribute extraction means for extracting attributes whose easiness of change differs depending on the attribute value of the subject.
The investigation device according to the present invention is characterized by comprising attribute value detection means for detecting a characteristic value of the subject attribute for each presence / absence of a change in the comprehensive evaluation.

In the investigation apparatus according to the present invention, for each subject whose overall evaluation has changed between the investigation stages, for each stimulus presented to the subject, an evaluation based on the stimulus presented to the subject was performed. Comprehensive evaluation change amount calculating means for calculating a change amount of the comprehensive evaluation is provided for each subject attribute and for each subject attribute that has changed in the comprehensive evaluation.
In the investigation apparatus according to the present invention, the comprehensive evaluation change amount calculating means calculates the change amount of the comprehensive evaluation for each direction of change of the comprehensive evaluation.

The investigation apparatus according to the present invention is characterized by comprising output data generation means for generating output data for comparing the analysis data related to a plurality of investigation objects for each of the investigation stages.
In the survey apparatus according to the present invention, the output data generation means generates output data that expresses the relationship among the survey target, the survey stage, and the partial evaluation in a multidimensional manner.
In the survey device according to the present invention, the output data generation means includes a cumulative contribution rate, column coordinate values, and row coordinates, which are analysis data of three types of quantification targeting a set of a survey target and a survey stage and partial evaluation. Based on the value, output data for generating a multidimensional arrangement of a set of a survey target and a survey stage and a partial evaluation is generated.
In the survey apparatus according to the present invention, the output data generation means includes a cumulative contribution rate, a sequence of analysis data of three types of quantification targeting a set of a survey target and a survey stage, a partial evaluation, and a subject attribute. Based on the coordinate value and the line coordinate value, output data for generating a multidimensional arrangement of a set of a survey target and a survey stage, a partial evaluation, and a subject attribute is generated.

The computer program according to the present invention provides a survey stage for presenting information on a survey target to obtain information on the survey target for a group of subjects, and a new stimulus based on the evaluation acquired in the survey stage. Repetitively performing presentation to a subject at a stage, storing information regarding the acquired evaluation in a storage means, and generating analysis data regarding a change in the evaluation between the investigation stages. A computer program for execution.
As a result, the above-described investigation device can be realized using a computer.

  According to the present invention, there is an effect that it is possible to contribute to grasping how the attitude of the subject group with respect to the investigation object is changed by the stimulus.

It is a block diagram which shows the structure of the investigation apparatus 1 which concerns on one Embodiment of this invention. It is a flowchart which shows the whole procedure of the investigation method which concerns on one Embodiment of this invention. It is an Example of the processing data which concerns on this invention. It is a structural example of the analysis data which concerns on Example 1 of this invention. It is a structural example of the analysis data which concerns on Example 2 of this invention. It is a structural example of the analysis data which concerns on Example 3 of this invention. It is a structural example of the analysis data which concerns on Example 4 of this invention. It is a structural example of the analysis data which concerns on Example 5 of this invention. It is an example of a structure of the analysis data which concerns on Example 6 of this invention. It is a structural example of the analysis data which concerns on Example 7 of this invention. It is an example of a structure of the analysis data which concerns on Example 8 of this invention. It is a structural example of the analysis data which concerns on Example 9 of this invention. It is a structural example of the analysis data which concerns on Example 10 of this invention. It is another structural example of the analysis data which concerns on Example 10 of this invention. It is a structural example of the decision tree based on Example 10 of this invention. It is a structural example of the analysis data which concerns on Example 11 of this invention. It is another structural example of the analysis data based on Example 11 of this invention. It is a structural example of the decision tree based on Example 11 of this invention. It is a structural example of the output data which concerns on Example 13 of this invention. It is another example of composition of output data concerning Example 13 of the present invention. It is an example of composition of processing data concerning Example 14 of the present invention. It is an example of the frequency count table which concerns on Example 14 of this invention. It is an example of the column coordinate value which concerns on Example 14 of this invention. It is an example of the line coordinate value which concerns on Example 14 of this invention. It is an example of the two-dimensional map which concerns on Example 14 of this invention. It is another example of the two-dimensional map which concerns on Example 14 of this invention. It is another example of the two-dimensional map which concerns on Example 14 of this invention. It is an example of a multiple cross table.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of an investigation apparatus 1 according to an embodiment of the present invention. In FIG. 1, a survey apparatus 1 includes a survey execution module 2, a survey response storage module 3, a survey result processing module 4, a survey result analysis module 5, an output data generation module 6, a raw data storage unit 7, and a processing data storage unit 8. An analysis data storage unit 9 is provided.

  The survey execution module 2 repeatedly performs a survey stage in which information related to the survey target is presented to the subject group and evaluation for the survey target is acquired. At this time, each time the investigation execution module 2 repeats the investigation stage, the investigation execution module 2 presents the stimulus based on the evaluation acquired in the past investigation stage in the repetition to the subject at the new investigation stage. As a result, when the testimony of the subject with respect to the survey target is changed by the presented stimulus, it can be expected that the change in the testimony is reflected in the evaluation of the subject with respect to the survey target.

  The survey response accumulation module 3 stores information related to evaluation acquired at the survey stage performed by the survey execution module 2 in the raw data storage unit 7. The raw data storage unit 7 accumulates information related to the evaluation acquired at each investigation stage. The survey result processing module 4 processes the information regarding the evaluation stored in the raw data storage unit 7 to generate processed data. The investigation result processing module 4 stores the generated processing data in the processing data storage unit 8. The machining data storage unit 8 stores machining data.

  The survey result analysis module 5 uses the processing data in the processing data storage unit 8 to generate analysis data regarding changes in evaluation between survey stages. The investigation result analysis module 5 stores the generated analysis data in the analysis data storage unit 9. The analysis data storage unit 9 stores analysis data. The output data generation module 6 generates output data using the analysis data in the analysis data storage unit 9.

  Next, the overall operation of the investigation apparatus 1 shown in FIG. 1 will be described with reference to FIG. FIG. 2 is a flowchart showing the overall procedure of the investigation method according to the present embodiment. In the following description, the investigation stage is simply referred to as a stage.

  In FIG. 2, in step S <b> 1, as a first stage, the survey execution module 2 presents a questionnaire regarding the survey target to the subject group and obtains an answer. For the presentation and answering of the questionnaire, for example, a website on the Internet may be used, or an e-mail may be used. Then, the survey response accumulation module 3 stores the first stage response in the raw data storage unit 7 as it is. Then, the survey result processing module 4 processes the answer of the first stage in the raw data storage unit 7 and stores the processed data in the processed data storage unit 8.

  Next, in step S2, the survey execution module 2 updates the stage. This increases the stage number by one. For example, the process moves from the first stage to the second stage. Hereinafter, it is set as the nth stage.

  Next, in step S3, the survey execution module 2 presents the survey target to the subject group. The survey target is presented using data of a type (image, sound, text, etc.) corresponding to the survey target. Next, in step S4, the survey execution module 2 presents a stimulus to the subject. The survey execution module 2 generates a stimulus using the answer in the raw data storage unit 7.

  Next, in step S5, the survey execution module 2 presents a questionnaire regarding the survey target and obtains an answer. Next, in step S6, the survey response accumulation module 3 stores the n-th stage response in the raw data storage unit 7 as it is. Next, in step S <b> 7, the survey result processing module 4 processes the answer of the nth stage in the raw data storage unit 7 and stores the processed data in the processed data storage unit 8.

  Next, in step S8, the survey execution module 2 determines the end of the stage. As a result, when the stage is finished, the process proceeds to step S9, and when the stage is not finished yet, the process returns to step S2.

  In step S <b> 9, the survey result analysis module 5 uses the processing data in the processing data storage unit 8 to generate analysis data regarding changes in evaluation between survey stages. Then, the survey result analysis module 5 stores the generated analysis data in the analysis data storage unit 9. Thereafter, the output data generation module 6 generates output data using the analysis data in the analysis data storage unit 9. The output data can be displayed on the screen of a display device such as a CRT (Cathode Ray Tube) or a liquid crystal display device, printed on a printing device, transmitted using a communication line such as the Internet, or the like. It can be stored in a portable recording medium.

  Next, analysis data generation processing performed by the survey result analysis module 5 and output data generation processing performed by the output data generation module 6 according to the present embodiment will be described with some examples.

First, items common to the embodiments will be described.
[Survey target]
One or a plurality of investigation targets can be arbitrarily set. Examples of survey targets include products, services, buildings, organizations, and the like. In the present embodiment, three companies (Company A, Company B, Company C) that handle similar products are surveyed.

[Evaluation]
There are two types of evaluation for survey subjects: comprehensive evaluation and partial evaluation. Comprehensive evaluation is a comprehensive evaluation of the survey target. Generally, comprehensive evaluation is expressed numerically. The partial evaluation is an elemental evaluation that constitutes the comprehensive evaluation. For partial evaluation, numerical values, text, images, and the like can be used. Comprehensive evaluation according to the present embodiment is the degree of preference (5 levels; 5 points: very fond, 4 points: fond, 3 points: neither can be said, 2) Point: dislike, 1 point: very dislike). The subject selects one of the five degrees of preference for each of the companies to be surveyed (Company A, Company B, Company C) and answers the questionnaire. In the first stage, the partial evaluation according to the present embodiment is a free description for each company (company A, company B, company C) in the first stage, and in the second and subsequent stages, the free description or the answer obtained in the first stage is the same. Select from (evaluation terms). For each company to be surveyed (Company A, Company B, Company C), the subject writes the evaluation in text and answers the questionnaire in the first stage, and writes the evaluation in text in the second and subsequent stages, or the questionnaire Answer the questionnaire by selecting from the evaluation terms presented in. In the present embodiment, it is clearly shown in the questionnaire that the theme of the questionnaire is “image survey of each company to be investigated (Company A, Company B, Company C)”. For this reason, a test subject answers the word (image word) showing the image with respect to each company (A company, B company, and C company) of investigation object as partial evaluation.

  The survey result processing module 4 performs text mining on the free description part in the questionnaire response, and collects notation fluctuations, synonyms and synonyms. As a result, partial evaluation based on free description can be used as an evaluation term. When the partial evaluation is an image, the image of each partial evaluation is digitized with a feature amount such as hue, and similar images are collected.

[Subject attributes]
The subject attributes are demographic attributes and information passivity. Demographic attributes include gender, age, occupation, residential area, family composition, and the like. There are three types of information passivity: a type that does not exchange information, a type that teaches information, and a type that teaches information. Include subject attributes as questionnaire questions. A test subject answers a demographic attribute and information passiveness by a questionnaire.

[stimulation]
For stimulation, evaluation (overall evaluation or partial evaluation) or subject attributes included in the questionnaire response can be used. In this embodiment, a stimulus is generated for each subject. The stimulus to be presented to a certain subject is the overall evaluation and partial evaluation performed by one other subject (stimulating user) in one stage, and the attribute of the stimulating user. Further, which stage of which stimulating user is used for the stimulation is randomly selected.

  In this embodiment, the data for presenting the survey target and the questionnaire regarding the survey target are information regarding the survey target. Moreover, the stimulus presented to the test subject identifier (subject ID), the questionnaire response (total evaluation and partial evaluation of the survey target, and the test subject attribute) and the test subject (questionnaire respondent) is information related to the evaluation.

  The number of subjects is set to a certain number (for example, 100 people) for each company (A company, B company, C company) to be investigated. The number of stages is four (from the first stage to the fourth stage).

  FIG. 3 is a configuration example of machining data according to the present embodiment. The survey result processing module 4 generates processing data from the questionnaire response in the raw data storage unit 7. The questionnaire response has an evaluation (comprehensive evaluation and partial evaluation) for the survey target and an attribute of the subject. As illustrated in FIG. 3, the processed data includes information regarding a survey target, a subject, a stage, a stimulus presented to the subject, and a subject's evaluation. Specifically, the processed data was given the name (subject name) of the survey target, the subject ID, the subject attribute (subject attribute), the stage number, and the stimulus presented to the subject (given stimulus). It has a set of a subject ID of a subject (stimulation user) who performed an evaluation based on the stimulus, an attribute of the stimulus user, a comprehensive evaluation of the subject, and a partial evaluation of the subject. The survey result analysis module 5 generates analysis data using the processed data. In this embodiment, the stimulus user's comprehensive evaluation and partial evaluation, and the stimulus user's attributes are the given stimuli.

  The contents of the analysis data shown in the following examples are for convenience of explanation and are provisional contents for explanation.

  FIG. 4 is a configuration example of analysis data according to the first embodiment of the present invention. The survey result analysis module 5 totalizes the comprehensive evaluation for each stage for each survey target. As shown in FIG. 4, the survey result analysis module 5 changes between the stages of the comprehensive evaluation of the subject group regarding one survey target (the company A is the example of FIG. 4), as illustrated in FIG. 4. Creates analysis data representing the change in the ratio of each value of the overall evaluation. From this analysis data, it is possible to grasp how the comprehensive evaluation of the subject group changes due to stimulation. In the example of FIG. 4, it can be seen that the ratio of likes (4, 5) increases in the overall evaluation of company A as the stage progresses.

  FIG. 5 is a configuration example of analysis data according to the second embodiment of the present invention. The survey result analysis module 5 totals partial evaluations for each stage for each survey target. As shown in FIG. 5, the survey result analysis module 5 changes between the stages of the partial evaluation of the subject group regarding one survey target (the company A is the example of FIG. 5), as illustrated in FIG. 5. Creates analysis data representing the change in the ranking of the number of evaluation terms for partial evaluation). From this analysis data, it is possible to grasp how the partial evaluation of the subject group changes due to stimulation. In the example of FIG. 5, it can be seen that “orange” and “music” appear in the upper rank in the partial evaluation of company A in common at all stages.

FIG. 6 is a configuration example of analysis data according to the third embodiment of the present invention. The survey result analysis module 5 calculates the importance of partial evaluation for each stage for each survey target. The importance is calculated by the following formula.
Importance = Number of cases at the stage x (Number of cases at the stage / Total number of cases at all stages)
As shown in FIG. 6, the survey result analysis module 5 changes between the stages of the partial evaluation of the subject group regarding one survey target (the company A is the example of FIG. 6) from the calculation result (example of FIG. 6). Creates analysis data representing the change in importance ranking of each evaluation term in the partial evaluation. From this analysis data, it is possible to grasp how the partial evaluation of the subject group changes due to stimulation. In the example of FIG. 6, as shown by the underline in the partial evaluation of company A, characteristic image words appear at the top, such as “young” in the first stage and “fashionable” in the fourth stage. I understand that.

  FIG. 7 is a configuration example of analysis data according to the fourth embodiment of the present invention. The survey result analysis module 5 totals partial evaluations for each survey stage and for each comprehensive evaluation for each survey target. As illustrated in FIG. 7, the survey result analysis module 5, as illustrated in FIG. 7, classifies one survey target (Company A is the example of FIG. 7) by comprehensive evaluation (the example of FIG. 7 is “4: Analytical data representing the change between the stages of partial evaluation of the subject group (like “like” and “2: dislike”) (the example of FIG. 7 is the change in the ranking of the number of each evaluation term in the partial evaluation) is created. From this analysis data, it is possible to grasp how the partial evaluation of the subject group changes due to stimulation for each comprehensive evaluation. In the example of FIG. 7, it can be seen that “orange” and “music” appear in the upper rank in common for all stages, regardless of likes and dislikes of the partial evaluation of company A.

  FIG. 8 is a configuration example of analysis data according to the fifth embodiment of the present invention. The survey result analysis module 5 calculates the importance of partial evaluation for each stage and for each comprehensive evaluation for each survey target. As shown in FIG. 8, the survey result analysis module 5, as illustrated in FIG. 8, relates to one survey target (the example of FIG. 8 is company A) by comprehensive evaluation (the example of FIG. 7 is “4: Analytical data representing changes between the stages of partial evaluation of the subject group (like “like” and “2: dislike”) (change in importance ranking of each evaluation term in the partial evaluation) is created. From this analysis data, it is possible to grasp how the partial evaluation of the subject group changes due to stimulation for each comprehensive evaluation. In the example of FIG. 8, as shown by the underline in the partial evaluation of company A, it can be seen that the subject who answered that he / she likes changes from “young” to “urban” or “fashionable” as the stage progresses. On the other hand, it can be seen that the subject who answered that he disliked had the image of “2nd” regardless of the stage.

  FIG. 9 is a configuration example of analysis data according to the sixth embodiment of the present invention. The survey result analysis module 5 discriminates subjects who have changed in overall evaluation and subjects who have not changed between stages for each survey target, and totals the stimuli presented to the subjects (given stimuli). As shown in FIG. 9, the survey result analysis module 5 compares the stimuli according to whether or not there is a change in the overall evaluation for one survey target (the company A is the example of FIG. 9), as illustrated in FIG. 9. The example of (2) creates analysis data representing the comparison of the ranking of the number of each stimulus. From this analysis data, it is possible to grasp the stimulus that has contributed to the change in the comprehensive evaluation and the stimulus that does not contribute to the change in the comprehensive evaluation.

  FIG. 10 is a configuration example of analysis data according to the seventh embodiment of the present invention. The survey result analysis module 5 calculates the importance of the stimulus (given stimulus) presented to the subject by distinguishing the subject who has changed the overall evaluation between the stages and the subject who has not changed for each survey target. The calculation formula for the importance of the stimulus is the same as the calculation formula for the importance of the partial evaluation. Based on the calculation result, the survey result analysis module 5 compares the importance of the stimulus according to whether or not there is a change in the overall evaluation regarding one survey target (Company A in the example of FIG. 10) as illustrated in FIG. Analytical data representing the comparison of importance rankings of the respective stimuli is created. From this analysis data, it is possible to grasp the importance of the stimulus that contributed to the change in the comprehensive evaluation and the importance of the stimulus that does not contribute to the change in the comprehensive evaluation. In the example of FIG. 10, it can be seen that for Company A, stimuli such as “Young”, “No. 2”, “Fashion”, “A (Trademark of Company A)” affect the change.

  FIG. 11 is a configuration example of analysis data according to the eighth embodiment of the present invention. The survey result analysis module 5 totals the stimuli (given stimuli) presented to the subject whose overall evaluation has changed between stages for each survey target, according to the direction of the change. As shown in FIG. 11, the survey result analysis module 5 determines, based on the result of the aggregation, the direction of the overall evaluation for one survey target (the example in FIG. 11 is Company A) (the example in FIG. Comparison of stimuli between the change in the direction of evaluation (4, 5) (+ direction) and the change in the direction of dislike (2, 1) (-direction) (the example in FIG. Analytical data representing comparison of number ranking) is created. From this analysis data, it is possible to grasp the stimulus that contributed to the change in the comprehensive evaluation in the + direction and the stimulus that contributed to the change in the comprehensive evaluation in the-direction. In addition, the change made into a total object can also make not only the change according to the direction of comprehensive evaluation but the change according to the direction of partial evaluation into a total object. For example, in the image evaluation as the partial evaluation, a change according to the direction of the image is set as an aggregation target. Examples of the image direction include “refreshing direction” and “heavy system direction”.

  FIG. 12 is a configuration example of analysis data according to the ninth embodiment of the present invention. The survey result analysis module 5 calculates the importance of the stimulus (given stimulus) presented to the subject whose overall evaluation has changed between stages for each survey target, for each change direction. As shown in FIG. 12, the survey result analysis module 5 determines, based on the calculation result, the direction of the overall evaluation for one survey target (company A in the example of FIG. 12). Comparison of the importance of stimulation between the change in the direction of evaluation (4, 5) (+ direction) and the change in the direction of dislike (2, 1) (-direction) (the example of FIG. 12 is Analytical data representing the comparison of the importance ranking of each stimulus is created. From this analysis data, it is possible to grasp the importance of the stimulus that contributed to the change in the comprehensive evaluation in the positive direction and the importance of the stimulus that contributed to the change in the comprehensive evaluation in the negative direction. In the example of FIG. 12, regarding “Company A”, “Young” and “Fashion” affect the change in the + direction, and “No. 2” and “A (trademark of Company A)” affect the change in the − direction. You can see that

According to the above-described Examples 1 to 9, it is possible to analyze the comprehensive evaluation or partial evaluation of the subject with respect to the investigation target, or the given stimulus.
If the partial evaluation is not a free description text but an individual evaluation score presented in a questionnaire, the decision tree can be used to easily stimulate the presumed cause of a change in the overall evaluation. Can be extracted. In addition, even if the partial evaluation is free description text, categorization is performed, and if the frequency of aggregation is large, the decision tree is used so that the stimulus that is presumed to have caused the change in the overall evaluation can be easily achieved. It can be expected to extract.

  If the partial evaluation is an image, image analysis can be performed to create an image attribute ranking. Also in this case, by using the decision tree, it can be expected that a stimulus that is presumed to have caused a change in the comprehensive evaluation is easily extracted from the image attribute item.

  FIG. 13 is a configuration example of analysis data according to the tenth embodiment of the present invention. The survey result analysis module 5 discriminates subjects who have changed in the overall evaluation between the stages and subjects who have not changed for each survey target, and performs evaluation based on the stimulus presented to the subject (given stimulus). The attributes of the performed subjects (stimulating users) are tabulated. As shown in FIG. 13, the survey result analysis module 5 compares stimulus user attributes according to whether or not there is a change in the overall evaluation regarding one survey target (the company A is an example of FIG. 13). The example of FIG. 13 creates analysis data representing the value of each stimulus user attribute (comparison of the number and ratio for each gender value (male, female), age value (20s, 30s, 40s)). . From this analysis data, it is possible to grasp the stimulus user attributes that contribute to the change in the comprehensive evaluation and the stimulus user attributes that do not contribute to the change in the comprehensive evaluation.

  Moreover, the survey result analysis module 5 totals the stimulus user attributes for each direction of change in the comprehensive evaluation. As shown in FIG. 14, the survey result analysis module 5 determines, based on the result of the aggregation, the direction of the overall evaluation for one survey target (the example of FIG. 14 is company A). Comparison of stimulation user attributes of change in direction (+ direction) of likes (4, 5) of evaluation and change in direction (− direction) of dislikes (2, 1) (in the example of FIG. Analytical data representing the stimulation user attribute values (comparison of the number and ratio for each sex value (male, female)) is created. From this analysis data, it is possible to grasp the stimulus user attribute that contributed to the change in the comprehensive evaluation in the + direction and the stimulus user attribute that contributed to the change in the comprehensive evaluation in the-direction. In the example of FIG. 14, it can be understood that there are many women who are easily stimulated to change in the + direction and many men who are easily stimulated to change in the − direction.

  When there are many types of subject attributes, analysis data can be easily generated by creating a decision tree as illustrated in FIG. Alternatively, a multiple cross table may be created.

  FIG. 16 is a configuration example of analysis data according to Embodiment 11 of the present invention. The survey result analysis module 5 distinguishes between subjects whose overall evaluation has changed between stages and subjects who have not changed for each survey target, and tabulates subject attributes. As shown in FIG. 16, the survey result analysis module 5 compares the subject attributes according to whether or not there is a change in the overall evaluation regarding one survey target (the company A is the example of FIG. 16). 16 examples are the number and ratio of each subject attribute value (gender value (male, female), information passivity value (type not to exchange information, type to receive information, type to teach information)) Analytical data representing (comparison) is created. From this analysis data, it is possible to grasp the value of the subject attribute whose comprehensive evaluation is likely to change. Also, it is possible to extract subject attributes that make it easy to determine the ease of change in the overall evaluation. In the example of FIG. 16, it can be seen that the ease of change differs depending on the value of information passivity. As a result, the survey result analysis module 5 extracts the information passivity as a subject attribute that makes it easy to determine the ease of change in the overall evaluation.

  In addition, the survey result analysis module 5 detects characteristic values of the subject attribute for each presence / absence of a change in the comprehensive evaluation. As the subject attribute to be detected, those extracted as subject attributes that are easy to determine the ease of change in the overall evaluation may be used. As illustrated in FIG. 17, the survey result analysis module 5 uses the detection result to determine the characteristic of the subject attribute according to whether or not there is a change in the overall evaluation regarding one survey target (the example of FIG. 17 is company A). Analysis data representing comparison of values (values indicated by underline in FIG. 17) (comparison of characteristic values of information passivity in the example of FIG. 17) is created. From this analysis data, it is possible to grasp the value of the subject attribute that tends to change the comprehensive evaluation and the value of the subject attribute that hardly changes the comprehensive evaluation. In the example of FIG. 17, it is understood that the person who teaches information easily changes the comprehensive evaluation, and the person who does not exchange information hardly changes the comprehensive evaluation.

  When there are many types of subject attributes, analysis data can be easily generated by creating a decision tree as illustrated in FIG. Alternatively, a multiple cross table may be created.

  According to Examples 10 to 11 described above, it is possible to analyze how the stimulus user attribute or the subject attribute affects the change in the comprehensive evaluation.

  The survey result analysis module 5 targets subjects whose overall evaluation has changed between the stages, and subjects who have performed evaluation based on the given stimulus for each stimulus (given stimulus) presented to the subject ( The change amount of the comprehensive evaluation is calculated for each attribute of the stimulation user) and for each attribute of the subject who has changed the comprehensive evaluation. Thereby, it is possible to compare the amount of change of each given stimulus, stimulus user attribute, and subject attribute. Furthermore, it is possible to grasp important factors related to changes in the overall evaluation. Further, the survey result analysis module 5 may calculate the change amount of the comprehensive evaluation for each direction of the change of the comprehensive evaluation.

  According to the twelfth embodiment, the applied stimulus, the stimulus user attribute, and the subject attribute can be handled in the same row, and the effects relating to the change in the comprehensive evaluation can be compared. And the effect that the important factor regarding the change of comprehensive evaluation can be grasped | ascertained by the comparison is acquired.

  FIG. 19 is a configuration example of output data according to the thirteenth embodiment of the present invention. The output data generation module 6 generates output data for comparing analysis data related to a plurality of investigation targets for each stage. In the example of FIG. 19, the output data is displayed as a graph by arranging the average values of the scores of the comprehensive evaluation collected for each stage related to each company to be investigated (Company A, Company B, Company C). Thereby, the average value of the score of the comprehensive evaluation total | counted for every stage can be compared about each company (A company, B company, C company) of investigation object. And the difference in the change of comprehensive evaluation between the stages of each company (company A, company B, company C) to be investigated can be grasped. In addition, although the average value of the score of comprehensive evaluation was used in the example of FIG. 19, you may use the total value of the score of comprehensive evaluation.

  FIG. 20 is another configuration example of output data according to Embodiment 13 of the present invention. In the example of FIG. 20, the output data is displayed in a tabular form with partial evaluation rankings created for each stage related to each company (A company, B company, and C company) to be investigated. Thereby, it is possible to compare the rankings of the partial evaluations created for each stage for each company to be investigated (Company A, Company B, Company C). And the difference of the partial evaluation between the stages of each company (A company, B company, and C company) of investigation object can be grasped. In the example of FIG. 19, the ranking is output. However, a certain number of partial evaluations having a high degree of importance are selected and listed for each stage related to each company (A company, B company, and C company) to be surveyed. You may make it output in a format. Further, as the partial evaluation, an image can be used in addition to text such as an image word. Alternatively, image features may be extracted from images answered as partial evaluations to aggregate similar images, and a ranking may be created for the aggregated images (evaluation images).

  According to the thirteenth embodiment, the difference between the comprehensive evaluation or the partial evaluation can be analyzed by the set of the investigation object and the stage.

The output data generation module 6 generates output data that expresses the relationship between the investigation target, the stage, and the partial evaluation in two dimensions. This will be specifically described below.
FIG. 21 is a configuration example of machining data according to the fourteenth embodiment of the present invention. In the example of FIG. 21, the processed data includes the subject ID, the name of the survey object, the stage number, and the partial evaluation terms (in the example of FIG. 21, “urban”, “fashionable”, “large company”, “business”). , “Animal”, “humor”, etc.) occurrences (1) or no (blank).

  The survey result analysis module 5 uses the processed data to create a frequency tabulation table of a pair of survey target and stage (survey target-stage) versus partial evaluation as shown in FIG. Next, the survey result analysis module 5 analyzes the “survey target-stage” versus partial evaluation using the quantification type 3 for the frequency summary table of “survey target-stage” versus partial evaluation. At this time, the investigation result analysis module 5 determines the number of dimensions while examining the contribution rate (cumulative contribution rate) of the eigenvalue. In this example, the number of dimensions is set to 2 based on the high cumulative contribution rate. Thus, in order to create a two-dimensional map, row coordinate values targeting “investigation target-stage” and column coordinate values targeting partial evaluation are calculated. FIG. 23 is an example of column coordinate values targeted for partial evaluation. FIG. 24 is an example of row coordinate values targeting “investigation target-stage”. Column coordinate values targeted for partial evaluation are two-dimensional coordinate values (component 1, component 2). The row coordinate value targeting “investigation target-stage” is a two-dimensional coordinate value (component 1, component 2).

  The output data generation module 6 includes a column coordinate value (two-dimensional coordinate value (component 1, component 2)) targeted for partial evaluation, and a row coordinate value (two-dimensional coordinate value ( Component 1 and component 2)) are used to generate output data for two-dimensionally arranging “investigation target-stage” and evaluation terms for partial evaluation. FIG. 25 is an example of a two-dimensional map in which “survey target-stage” and evaluation terms for partial evaluation are two-dimensionally arranged. In FIG. 25, component 1 is a first dimension value and component 2 is a second dimension value. According to this two-dimensional map, while grasping the relative positioning (competitive relationship) between the investigation objects, it is possible to grasp the change in the positioning as the stage progresses. Furthermore, it is possible to grasp the change in partial evaluation accompanying the progress of the stage for each survey target.

  FIG. 26 is another example of a two-dimensional map according to Embodiment 14 of the present invention. The example of FIG. 26 is an example of a two-dimensional map when an image is also used for partial evaluation. An evaluation image is arranged on the two-dimensional map.

FIG. 27 is another example of a two-dimensional map according to Embodiment 14 of the present invention. In the example of FIG. 27, “investigation target—stage”, partial evaluation (evaluation term, evaluation image), and subject attributes are arranged two-dimensionally. For this reason, the survey result analysis module 5 performs three types of quantifications on “survey target-stage”, partial evaluation (evaluation terms, evaluation images), and subject attributes, and calculates a cumulative contribution rate. .
Although the two-dimensional map is created in the fourteenth embodiment, the number of dimensions can be arbitrarily selected based on the cumulative contribution rate, and may be three or more.

As mentioned above, although embodiment of this invention was explained in full detail with reference to drawings, the specific structure is not restricted to this embodiment, The design change etc. of the range which does not deviate from the summary of this invention are included.
For example, as the data format handled in the analysis data generation processing according to the present embodiment, the processing data itself such as “item category type” and “0-1 type” is used, or the factor to be focused on is shown in FIG. A multiple cross table as shown in Fig. 1 can be created and used.

  As described above, according to the present embodiment, it is possible to efficiently simulate how the evaluation of the subject group changes due to the stimulus, and further, what factor causes the evaluation to change with respect to the investigation target. it can. Thereby, the special effect that it becomes possible to fully analyze the influence which CGM has on consumers is obtained.

Further, according to the present embodiment, the following effects can be obtained.
(1) It is possible to grasp the actual state of changes in attitudes of subjects to the survey target. For example, it is possible to find the convergence direction of evaluation for a product. Specifically, the convergence direction of the comprehensive evaluation can be found such that the comprehensive evaluation (like or dislike) increases or decreases according to the stimulus. Or, it is possible to discover the convergence direction of partial evaluation, such as partial evaluation (image word) changes according to the stimulus (the image word related to design is mainly the image word related to price). .

(2) It is possible to find a stimulus (past evaluation, stimulus user attribute) that triggered the change in attitude of the subject to the survey target. For example, it is possible to discover what past evaluations contribute to a good evaluation for a product and what past evaluations contribute to a bad evaluation for a product. In addition, it is possible to discover what kind of attribute a stimulus from a subject contributes to a good evaluation for a product and what kind of attribute a subject's stimulus contributes to a bad evaluation for a product. This can be used, for example, to simulate changes in consumer attitudes toward products by intentionally putting a message that a company wants to send.

(3) It is possible to discover what kind of attributes the subject is likely to change the attitude toward the survey target.
(4) It is possible to compare the content of the subject's attitude change, the stimulus that triggers the attitude change, the characteristics of the subject who is likely to cause the attitude change, and the like among a plurality of survey subjects.

  Note that the survey apparatus 1 according to the present embodiment may be realized by dedicated hardware, or is configured by a computer system such as a personal computer, and functions of each unit of the survey apparatus 1 shown in FIG. The function may be realized by executing a program for realizing the above.

Further, it is assumed that an input device, an output device, etc. (none of which are shown) are connected to the investigation device 1 as peripheral devices. Here, the input device refers to an input device such as a keyboard and a mouse, a reading device that reads data from a recording medium, and the like. Examples of the output device include a display device such as a CRT (Cathode Ray Tube) and a liquid crystal display device, a recording device for a recording medium, and a printing device.
The peripheral device may be directly connected to the investigation apparatus 1 or may be connected via a communication line.

2 is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into the computer system and executed, thereby performing an investigation process. May be. Here, the “computer system” may include an OS and hardware such as peripheral devices.
Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used.
“Computer-readable recording medium” refers to a flexible disk, a magneto-optical disk, a ROM, a writable nonvolatile memory such as a flash memory, a portable medium such as a DVD (Digital Versatile Disk), and a built-in computer system. A storage device such as a hard disk.

Further, the “computer-readable recording medium” means a volatile memory (for example, DRAM (Dynamic DRAM) in a computer system that becomes a server or a client when a program is transmitted through a network such as the Internet or a communication line such as a telephone line. Random Access Memory)), etc., which hold programs for a certain period of time.
The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line.
The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

DESCRIPTION OF SYMBOLS 1 ... Survey apparatus, 2 ... Survey execution module, 3 ... Survey response accumulation module, 4 ... Survey result processing module, 5 ... Survey result analysis module, 6 ... Output data generation module, 7 ... Raw data storage part, 8 ... Process data Storage unit, 9 ... analysis data storage unit

Claims (22)

For the test subject group, presenting information on the survey target and obtaining an evaluation for the survey target, while presenting the stimulus based on the evaluation acquired in the survey stage to the subject in the new survey stage, Repeated investigations,
Storage means for storing information relating to the acquired evaluation;
Analysis data generating means for generating analysis data relating to changes in the evaluation between the investigation stages;
An investigation device characterized by comprising:   The survey apparatus according to claim 1, further comprising a comprehensive evaluation totaling unit that totalizes total evaluation for each of the survey stages.   The survey apparatus according to claim 1, further comprising a partial evaluation totaling unit that totalizes partial evaluation for each of the survey stages.   The survey apparatus according to claim 1, further comprising a partial evaluation importance calculation unit that calculates the importance of partial evaluation for each survey stage.   The survey apparatus according to claim 1, further comprising a partial evaluation totaling unit that totalizes partial evaluations for each of the survey stages and for each comprehensive evaluation.   The survey apparatus according to claim 1, further comprising a partial evaluation importance calculation unit that calculates the importance of partial evaluation for each of the survey stages and for each comprehensive evaluation.   The survey according to claim 1, further comprising: a stimulus totaling unit that counts the stimuli presented to the subjects by distinguishing between subjects who have changed in the overall evaluation and subjects who have not changed between the survey stages. apparatus.   The stimulus importance calculation means for distinguishing between a subject whose overall evaluation has changed between the investigation stages and a subject who has not changed, and calculating the importance of the stimulus presented to the subject, 1. The investigation device according to 1.   The survey apparatus according to claim 7, wherein the stimulus totaling unit totals the stimuli presented to the subject whose overall evaluation has changed between the survey stages according to the direction of the change.   9. The investigation apparatus according to claim 8, wherein the stimulus importance degree calculation means calculates the importance degree of the stimulus presented to the subject whose overall evaluation has changed between the investigation stages for each change direction. .   An attribute totaling unit that aggregates the attributes of the subjects who performed the evaluation based on the stimulus presented to the subject by distinguishing the subjects who had a change in the overall evaluation between the survey stages and the subjects who did not change. The investigation device according to claim 1, wherein:   The survey apparatus according to claim 1, further comprising an attribute counting unit that counts subjects' attributes by distinguishing between subjects whose overall evaluation has changed and those who have not changed between the survey stages.   The survey apparatus according to claim 11, wherein the attribute totaling unit totalizes the attributes for each direction of change.   The survey apparatus according to claim 12, further comprising an attribute extraction unit configured to extract an attribute that is easily changed according to the attribute value of the subject.   The survey apparatus according to claim 14, further comprising attribute value detection means for detecting a characteristic value of the subject attribute for each presence / absence of a change in the comprehensive evaluation.   For subjects whose overall evaluation has changed between the survey stages, for each stimulus presented to the subject, for each attribute of the subject who performed the evaluation based on the stimulus presented to the subject, and for overall assessment The survey apparatus according to claim 1, further comprising: a comprehensive evaluation change amount calculating unit that calculates a change amount of the comprehensive evaluation for each attribute of the subject who has a problem.   The survey apparatus according to claim 16, wherein the comprehensive evaluation change amount calculation unit calculates the change amount of the comprehensive evaluation for each direction of change of the comprehensive evaluation.   The survey apparatus according to claim 1, further comprising output data generation means for generating output data for comparing the analysis data related to a plurality of survey targets for each of the survey stages.   19. The survey apparatus according to claim 18, wherein the output data generation means generates output data that expresses the relationship between the survey target, the survey stage, and the partial evaluation in a multidimensional manner.   The output data generation means, based on the cumulative contribution rate, the column coordinate value, and the row coordinate value, which are analysis data of the quantification type 3 for the set of the survey target and the survey stage and the partial evaluation, 20. The investigation apparatus according to claim 19, wherein output data for generating a multi-dimensional arrangement of a group of investigation stages and partial evaluation is generated.   The output data generation means is based on a cumulative contribution rate, column coordinate values, and row coordinate values, which are analysis data of three types of quantification targeting a set of a survey target and a survey stage, partial evaluation, and subject attributes. The survey apparatus according to claim 19, wherein output data for generating a multidimensional arrangement of a set of a survey target and a survey stage, a partial evaluation, and a subject attribute is generated. For the test subject group, presenting information on the survey target and obtaining an evaluation for the survey target, while presenting the stimulus based on the evaluation acquired in the survey stage to the subject in the new survey stage, Repeated steps,
Storing information on the acquired evaluation in a storage means;
Generating analytical data regarding changes in the assessment between the survey stages;
A computer program for causing a computer to execute.

Images