Analysis system

Abstract

Provided is an analysis system that contributes to the understanding of the influence of psychological characteristics on behaviors. An analysis system (11) is provided with: a storage unit (13) for storing analysis data including psychological characteristic data obtained by scoring psychological characteristics on the basis of question items; and an analysis unit (18) that performs cluster analysis on the analysis data and extracts feature items from the psychological characteristic data for each cluster.

Description

Technical Field

[0001] The present invention relates to an analysis system having a database for storing characteristic data obtained by scoring based on question items. Background Art

[0002] Patent Document 1 discloses a pressure sensitivity assessment paper. The pressure sensitivity assessment paper scores sensitivity to pressure based on questions. When scoring, the subject answers questions on the pressure sensitivity assessment paper. The evaluation score is set according to the performance of the answers.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Publication No. 2014-230553 Summary of the invention

[0006] Problem that the invention aims to solve

[0007] There are individual differences in people's behavior when purchasing goods and services, but it is not easy to understand the relationship between various individual differences and behavior. On the other hand, if people's psychological characteristics, such as sensitivity to stress, are classified according to characteristic items, it is possible to understand the impact of psychological characteristics, such as sensitivity to stress, on people's behavior.

[0008] The object of the present invention is to provide an analysis system that helps to understand the impact of psychological characteristics on behavior.

[0009] Solutions for solving problems

[0010] An analysis system according to one embodiment of the present invention comprises: a storage unit that stores analysis data including psychological characteristic data obtained by scoring psychological characteristics based on question items; and an analysis unit that performs cluster analysis on the analysis data and extracts feature items from the psychological characteristic data for each cluster.

[0011] Effects of the Invention

[0012] As described above, according to the present invention, it is possible to provide an analysis system that helps to understand the influence of psychological characteristics on behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a diagram schematically showing the configuration of a sensory behavior analysis system according to one embodiment of the present invention.

[0014] Figure 2 A diagram showing a specific example of a scree map.

[0015] Figure 3 It is a graph showing the principal component coefficient for each principal component.

[0016] Figure 4 is a graph showing the interval statistics by number of clusters.

[0017] Figure 5 This is a graph showing the centroid of the principal component for each cluster.

[0018] Figure 6 (A) is a graph showing the centroid of one principal component obtained when cluster analysis is performed with the number of factors "5". Figure 6 (B) is a graph showing the centroid of one principal component obtained when cluster analysis is performed with the number of factors "4".

[0019] Figure 7 (A) is a graph showing the centroid of one principal component obtained when cluster analysis is performed with the number of factors "5". Figure 7 (B) is a graph showing the centroid of one principal component obtained when cluster analysis is performed with the number of factors "4". DETAILED DESCRIPTION

[0020] Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

[0021] <Overall structure of sensory behavior analysis system>

[0022] Figure 1 The structure of a sensory behavior analysis system according to one embodiment of the present invention is schematically shown. The sensory behavior analysis system 11 includes a first storage unit 13 storing a first database 12, and a second storage unit 15 storing a second database 14. The first database 12 stores psychological characteristic data for determining scores of interoceptive sensations and stress-related indicators after scoring. The second database 14 stores purchase behavior data for determining a consumer's purchase behavior based on numerical values.

[0023] The psychological characteristic data includes numerical values ​​for determining scores of questionnaires such as "MAIA (Multidimensional Assessment of Interoceptive Sensation Awareness)", "BPQ", "JPSS", and "RS" for each ID. Each person is assigned an ID. The questions included in each questionnaire only need to be questions determined in the research field of psychology. MAIA presents 32 questions such as "When you are nervous, do you notice which part of your body is nervous?" and "Do you notice discomfort in your body?". For each question, a score is assigned according to [0 points = never] [1 point = almost never] [2 points = rarely] [3 points = occasionally] [4 points = often] [5 points = always]. MAIA can score body sensations and interoceptive sensations from eight aspects: "Attention (MAIA1)", "Not distracted (MAIA2)", "Not worried (MAIA3)", "Attention regulation (MAIA4)", "Emotional awareness (MAIA5)", "Self-regulation (MAIA6)", "Body listening (MAIA7)", and "Trust (MAIA8)". Questions are assigned to each aspect.

[0024] BPQ presents 46 questions, such as "frequent swallowing" and "the urge to cough and clear the throat". For each question, a score is assigned according to [1 point = never] [2 points = occasionally] [3 points = sometimes] [4 points = often] [5 points = always]. BPQ can score the awareness of interoceptive sensations and the sensitivity of interoceptive sensations. In BPQ, scoring can be implemented by dividing into "total score (BPQ_ALL)", "factor related to normal body perception (BPQ_BA)", "factor related to interoceptive sensation of the upper body (BPQ_Supra)", and "factor related to interoceptive sensation of the lower body (BPQ_Sub)". Questions are assigned to each factor.

[0025] Interoceptive senses are senses that capture the state of the body's interior. In contrast, exteroceptive senses are senses that capture information from outside the body, including vision, hearing, smell, taste, touch, and other senses.

[0026] JPSS presents 14 questions, such as "I was upset by unexpected events" and "I had trouble making important decisions" in the past month. For each question, you rate the degree of stress you feel on a five-point scale [0 to 4 points]. The degree of stress can be assessed based on the total score.

[0027] RS presents 14 questions, such as "I can rely on myself more than others" and "Sometimes I do things regardless of whether I want to or not." For each question, resilience (psychological resilience to stress) is scored on a seven-point scale [0 to 6 points]. Resilience can be assessed based on the total score.

[0028] In addition, when scoring psychological characteristics, scales related to self-cognition, scales related to general personality, scales related to motivation / desire, scales related to cognitive judgment tendencies, scales related to values, scales related to adaptation, scales related to emotions / mood, and scales related to depression and anxiety can also be used. More specifically, the following scales can be cited: Self-concept measurement scale, Self-stability scale, Self-recognition need scale, Concentration scale, Self-awareness scale, Narcissistic vulnerability scale, Unsatisfied self scale, Change motivation scale, Self-esteem scale, Self-acceptance measurement scale, Self-efficacy scale (SE scale), Acceptance / rejection scale, Assumed ability scale, Self-assertion / self-restraint cognitive scale, Self-identity model scale (IPS), Sense of fulfillment scale, Big Five personality measurement scale (NEO-Personality Inventory-R (NEO-PI-R)), Japanese version of the Five-Factor Personality Inventory (NEO-FFI (NEO Five-Factor Inventory)), Sensation-Seeking Scale, Stimulation Seeking Scale / Abstract Representation Item Version, Achievement-Related Motives Scale (Achievement-Related Motives Scale). Scale (ARMS), psychological needs scale, Japanese version of BIS / BAS scale, ambiguity tolerance scale, value scale, value orientation scale, "housing" psychological function measurement scale, empathy scale, and subjective well-being scale. By combining several scales, in addition to stress, it is possible to evaluate psychological characteristics such as personality, values, neophobia, empathy, cognitive ability, mood, emotion, depression, etc.

[0029] The purchase behavior data includes, for each ID, a numerical value for determining, for example, the "number of goods purchased throughout the year", a numerical value for determining the "types of goods purchased throughout the year", a numerical value for determining the "total amount of goods purchased throughout the year", a numerical value for determining the "average amount of unit prices", and a numerical value for determining the "index related to regular purchases". Each person is assigned an ID. Therefore, the purchase behavior of each person can be determined. The numerical value can be managed for each product. For example, the product can be identified based on the JAN code. The products can also be processed uniformly according to categories such as "yogurt", "chocolate", "fermented food", "dairy products", and "functional food".

[0030] The sensory behavior analysis system 11 includes a processing unit 16 connected to the first storage unit 13 and the second storage unit 15. The processing unit 16 includes: a data set generation unit (analysis data generation unit) 17, which combines the purchase behavior data obtained from the second database 14 with the psychological characteristic data obtained from the first database 12 to generate an analysis data set (analysis data); and an analysis unit 18, which performs cluster analysis on the analysis data set and extracts feature items from the purchase behavior data and the psychological characteristic data for each cluster. The data set generation unit 17 associates the psychological characteristic data with the purchase behavior data based on the ID. Therefore, an analysis data set is generated for each ID (each person). The analysis data set can be registered in the third database 21, for example. The third database 21 can be stored in the third storage unit 22 connected to the processing unit 16, for example.

[0031] When the analysis unit 18 performs cluster analysis on the analysis data set, for example, the K-means method is used. The analysis unit 18 generates the number of clusters based on the interval statistic. The interval statistic is calculated according to the cluster value "1", "2", "3" ... "k". When the interval statistic reaches the maximum value, the number of clusters can be determined.

[0032] The analysis unit 18 randomly sets the number of "representative points" of clusters. Calculate the distance between an analytical data set and the "representative point". Associate the analytical data set with the nearest "representative point". When all analytical data sets are associated with any "representative point", set the centroid of the cluster as a new "representative point". Calculate the distance between each analytical data set and the new "representative point". Associate the analytical data set with the new "representative point". Repeatedly set new "representative points" and when the centroid of the cluster is fixed, the cluster can be determined. When the cluster is repeatedly determined while changing the initial value of the "representative point", the cluster converges. The analysis unit 18 calculates the centroid for each cluster. The analysis unit 18 can implement the K-means method in the same way as Statistics and Machine Learning Toolbox from MathWorks, for example.

[0033] The operation processing unit 16 has an auxiliary analysis unit 23, which implements principal component analysis when the number of clusters exceeds a predetermined number of clusters when cluster analysis is performed on the analytical data set. Here, the number of clusters is set to "6", for example. When implementing the principal component analysis, the auxiliary analysis unit 23 calculates the optimal number of principal components. When performing this calculation, the auxiliary analysis unit 23 generates a scree plot. In the scree plot, the eigenvalues ​​are marked according to the number of factors of the principal component. When parallel analysis is applied to the scree plot, the number of factors of the principal component can be determined. The Kaiser criterion can also be used instead of parallel analysis. In the Kaiser criterion, the maximum number of factors is used among the eigenvalues ​​greater than 1. The number of principal components can be determined based on the characteristics of the product. The auxiliary analysis unit 23 calculates the principal component coefficients. The auxiliary analysis unit 23 can implement the principal component analysis in the same way as IBM's SPSS Statistics, for example.

[0034] The calculation processing unit 16 further includes a drawing unit 24, which generates graphic data that visually displays an index of at least one of the purchase behavior data and the psychological characteristic data for each cluster. The drawing unit 24 outputs the graphic data to a display panel 25, for example. The index includes, for example, the center of gravity of each cluster. In addition, the graphic data can visually display the principal component coefficient for each principal component.

[0035] The calculation processing unit 16 further includes a scoring unit 26 connected to the first storage unit 13 and used to register the psychological characteristic data in the first database 12; and a purchase quantification unit 27 connected to the second storage unit 15 and used to register the purchase behavior data in the second database 14. The purchase quantification unit 27 collects the purchase behavior data during a predetermined period. The purchase quantification unit 27 and the scoring unit 26 can be connected to an input device 29 such as a smartphone terminal, a tablet terminal, or a personal computer terminal via a network 28 such as the Internet.

[0036] The scoring unit 26 can obtain the feature original data from the input device 29 and process the feature original data into psychological characteristic data. The feature original data is used, for example, to determine a score for each question item. When obtaining the score, the input device 29 displays an input field for presenting the question items of the questionnaire on the screen of the display panel. The subject inputs an answer (score) for each question item from the keypad or keyboard. The scoring unit 26 can register the feature original data in the first storage unit 13.

[0037] The purchase digitization unit 27 can obtain the original purchase data of the determined period from the input device 29, and process the original purchase data into purchase behavior data. The original purchase data is used to determine the product code and the purchase amount, for example. In the case of obtaining the product code, the input device 29 has, for example, a barcode reader. When the subject purchases the product 31, the barcode 32 of the product 31 is read. The product can be determined for each barcode 32. In the case of obtaining the purchase amount, the input device 29 has a keypad and a keyboard. The subject enters the purchase amount, for example, manually. The purchase digitization unit 27 can register the original purchase data in the second storage unit 15. The purchase behavior data can be updated each time the product 31 is purchased.

[0038] The operation processing unit 16 can be constituted by a server computer, for example. The server computer executes an application program through a central processing unit (CPU), for example, to realize the function of the operation processing unit 16. When executing the application program, a large-capacity storage device for storing the application program, a memory for temporarily storing programs and data, etc. is connected to the CPU. When collecting the purchase raw data, the purchase digitization unit 27 can be implemented based on a separate application program separately from other functions. Similarly, when acquiring the feature raw data, the scoring unit 26 can be implemented based on a separate application program separately from other functions. The data set generation unit 17, the analysis unit 18, the auxiliary analysis unit 23 and the drawing unit 24, the third storage unit 22 and the display panel 25 can also be implemented by a personal computer. The first storage unit 13, the second storage unit 15, and the third storage unit 22 can also be respectively configured as, for example, a disk array device. In addition, the first storage unit 13, the second storage unit 15, and the third storage unit 22 can also be configured as one storage unit.

[0039] <Action>

[0040] Next, the operation of the sensory behavior analysis system 11 is described. The analysis unit 18 starts to perform cluster analysis on the analytical data set. The analysis unit 18 determines the number of clusters based on the interval statistic. When determining the number of clusters, the analysis unit 18 obtains a large number of analytical data sets. The analytical data set can be provided from the data set generation unit 17, or it can be obtained from the third database 21. In the analysis unit 18, as long as the interval statistic does not continue to diverge according to the increase in the number of clusters, the number of clusters can be uniquely determined. However, when the number of clusters exceeds the predetermined value "6", the analysis unit 18 terminates the cluster analysis.

[0041] If the number of clusters is less than "6", the analysis unit 18 continues the cluster analysis. The clusters are determined. The analysis unit 18 calculates the centroid for each cluster. The centroid is determined for each component factor included in the analysis data set. Each component factor can function as an indicator. The drawing unit 24 generates graphic data based on the calculated centroid. When the graphic data is provided to the display panel 25, the display panel 25 can visually display the indicator on the screen based on the graphic data. The characteristic items can be expressed for each cluster based on the displayed indicators. The observer observing the display can well interpret the characteristic items.

[0042] In this way, the analysis unit 18 performs cluster analysis on the analytical data set, and extracts characteristic items from the purchase behavior data and the psychological characteristic data for each cluster. By performing cluster analysis on the analytical data set, the relationship between the purchase behavior and the interoceptive feeling can be revealed. The influence of the psychological characteristics classified based on the interoceptive feeling on the purchase behavior can be understood.

[0043] If the number of clusters exceeds "6", the auxiliary analysis unit 23 performs principal component analysis to reduce the dimension. The auxiliary analysis unit 23 calculates the principal component coefficient for each principal component. The drawing unit 24 generates graphic data based on the calculated principal component coefficient. When the graphic data is provided to the display panel 25, the display panel 25 can visually display the principal component coefficient on the screen based on the graphic data. It is possible to express characteristic items for each principal component based on the displayed principal component coefficient. An observer observing the display can well interpret the characteristic items.

[0044] Next, the analysis unit 18 performs cluster analysis on the analysis data set based on the principal component. The analysis unit 18 determines the number of clusters based on the interval statistic. The number of factors of the cluster analysis is reduced accordingly with the principal component analysis. If the number of clusters is less than "6", the analysis unit 18 continues the cluster analysis. If the number of clusters exceeds "6", the analysis unit 18 reduces the number of factors. When reducing the number of factors, the analysis unit 18 performs factor analysis, for example. Since the number of factors is reduced, clusters can be formed well. Feature items can be highlighted for each cluster.

[0045] In the sensory behavior analysis system 11 of the present embodiment, purchase behavior data is collected during a predetermined period. Since the purchase behavior data is continuously collected during a specific period, the purchase behavior data can well reflect the purchase behavior of the commodity without being affected by the impulsive psychological state. The influence of the interoceptive feeling can be well analyzed. As the psychological characteristic data and the purchase behavior data increase, that is, as the number of subjects to be investigated increases, the deviation of the psychological characteristics and the purchase behavior can be eliminated for the whole, thereby achieving high-precision analysis.

[0046] In the present embodiment, when performing cluster analysis on the parsed data set, the K-means method is used. In the K-means method, the number of clusters is preset before forming the clusters. By restricting the number of clusters in this way, it is possible to classify psychological characteristics well based on interoceptive sensations. Characteristic matters can be highlighted for each cluster. When the number of clusters is large, the characteristic matters will be over-specified, and there is a risk that trivial characteristic matters will affect the formation of classification as noise.

[0047] The psychological characteristic data according to the present embodiment includes scores of indexes related to interoceptive sensations obtained based on the question items. Therefore, it is possible to grasp the influence of the psychological characteristics classified based on interoceptive sensations on the purchasing behavior. In addition, the psychological characteristic data includes scores of indexes related to stress obtained based on the question items. Therefore, it is possible to grasp the influence of the psychological characteristics classified based on stress on the purchasing behavior.

[0048] <Verification of actions>

[0049] The inventor of the present invention verified the operation of the sensory behavior analysis system 11. When performing the verification, purchase behavior data dedicated to yogurt was prepared. At this time, the purchase behavior data included values for determining "the number of yogurts purchased throughout the year (yogurt_p)", values for determining "the types of yogurts purchased throughout the year (yogurt_u)", values for determining "the total purchase amount of yogurts purchased throughout the year (yogurt_monetary)", values for determining "the average amount of the unit price (yogurt_mean)", and values for determining "the index related to regular purchases (yogurt_Hp)".

[0050] Here, the frequency distribution of the purchase behavior data is not a normal distribution but a long-tailed distribution (Japanese: a distribution with a long tail), so natural logarithm transformation was performed on all the data. When performing the natural logarithm transformation, "1" was added to all the data. The top 5% of the data was excluded from the analysis as outliers.

[0051] The answers to the questionnaires "MAIA", "BPQ", "JPSS" and "RS" were obtained from the corresponding data of the purchase behavior data. In MAIA, scores were determined for "attention (MAIA1)", "not being distracted (MAIA2)", "not worrying (MAIA3)", "attention regulation (MAIA4)", "emotional awareness (MAIA5)", "self-regulation (MAIA6)", "body listening (MAIA7)" and "trust (MAIA8)". In BPQ, scores were determined for "total score (BPQ_ALL)", "factors related to normal body perception (BPQ_BA)", "factors related to the interoceptive sensation of the upper body (BPQ_Sp)" and "factors related to the interoceptive sensation of the lower body (BPQ_Su)". In JPSS, the total score (JPSS) was determined. In RS, the total score (RS) was determined. The standard deviation was calculated for each factor for each ID. All IDs were standardized according to the robust z score. The standard deviation was judged as a deviation value at the level of 0.5% and the corresponding psychological characteristic data was excluded. As a result, a cluster analysis was performed on 6993 parsed datasets.

[0052] After cluster analysis was performed using the K-means method, the interval statistics did not converge even though the number of clusters exceeded 30, so dimensionality reduction was performed using principal component analysis. Figure 2 A scree plot was created as shown. After applying parallel analysis to the scree plot, the number of factors of the principal component was determined to be "4". A principal component analysis was performed based on the number of factors "4". The result is as follows Figure 3 The principal component coefficients were determined as shown. Based on the determined principal component coefficients, it was confirmed that [Component 1] reflects the influence of "Attention (MAIA1)", "Not distracted (MAIA2)", "Not worried (MAIA3)", "Attention regulation (MAIA4)", "Emotional awareness (MAIA5)", "Self-regulation (MAIA6)", "Body listening (MAIA7)", and "Trust (MAIA8)". It was confirmed that [Component 2] reflects the influence of the BPQ as a whole. It was confirmed that [Component 3] reflects the influence of "Number of yogurts purchased throughout the year (yogurt_p)", "Types of yogurts purchased throughout the year (yogurt_u)", "Total amount of yogurt purchased throughout the year (yogurt_monetary)", and "Index related to regular purchase (yogurt_Hp)". It was confirmed that [Component 4] reflects the influence of "Not worried (MAIA3)", JPSS, and RS.

[0053] Then, if Figure 4As shown in the figure, based on the derived principal components [component 1 = MAIA] [component 2 = BPQ] [component 3 = Yogurt] [component 4 = RS-JPSS], the interval statistics were calculated using the K-means method. As a result, the number of clusters was determined to be "5". The K-means method was implemented based on the number of clusters "5". As a result, Figure 5 As shown in the figure, the centroid of the principal component [MAIA][BPQ][Yogurt][RS-JPSS] is determined for each cluster [Cls1][Cls2][Cls3][Cls4][Cls5]. Figure 5 In , the clusters are arranged in ascending order based on [Component 3 = Yogurt].

[0054] The inventors examined the results of the cluster analysis. Figure 5 The center of gravity and Figure 3 The principal component coefficients of . An explanation was derived from the observation. The following psychological characteristics were found for [Cls2] who bought the most yogurt: worrying too much and having low stress tolerance, feeling stressed. For [Cls3] and [Cls4] who have high stress tolerance and no stress, it was found that the psychological characteristics of having a strong interoceptive feeling of MAIA are more supportive of buying yogurt than those with a weak interoceptive feeling.

[0055] Next, the inventors observed the influence of the number of factors of the principal component when performing cluster analysis. Figure 6 and Figure 7 As shown in Figure 1, based on the results of cluster analysis, the centroid of factor number "4" is compared with the centroid of factor number "5". In factor number "5", the number of clusters is set to "6". Figure 6 In the comparison, the centroids of [ingredient 3 = Yogurt] were compared. The clusters were arranged in ascending order of [ingredient 3 = Yogurt]. In the case of the number of factors "4", significant differences were well confirmed between all clusters. In the case of the number of factors "5", no sufficient significant differences were confirmed in [Cls4] and [Cls6]. Figure 7 , the centroids of [Component 1 = MAIA] were compared. The clusters were arranged in ascending order of [Component 3 = Yogurt]. In the case of the number of factors "4", no sufficiently significant difference was confirmed between [Cls1] and [Cls5]. Similarly, no sufficiently significant difference was confirmed between [Cls5] and [Cls2]. In contrast, in the case of the number of factors "5", no sufficiently significant difference was confirmed between [Cls1], [Cls3], and [Cls4]. Similarly, no sufficiently significant difference was confirmed between [Cls3], [CLs4], and [Cls5].

[0056] In addition, the sensory behavior analysis system 11 of the present embodiment may also include: a first storage unit 13 that stores a first database 12 for storing characteristic data obtained by scoring interoceptive sensations based on question items; a second storage unit 15 that stores a second database 14 for storing purchase behavior data for determining purchase behavior based on numerical values; a data set generation unit 17 that combines the characteristic data obtained from the first database 12 with the purchase behavior data obtained from the second database 14 to generate an analytical data set; and an analysis unit 18 that performs cluster analysis on the analytical data set to extract characteristic items from the purchase behavior data and the psychological characteristic data for each cluster. By performing cluster analysis on the analytical data set, the relationship between purchase behavior and interoceptive sensations can be revealed. The influence of the psychological characteristics classified based on the interoceptive sensations on the purchase behavior can be grasped.

[0057] <Notes>

[0058] (Supplementary Note 1) An analysis system comprising:

[0059] a storage unit storing analysis data including psychological characteristic data obtained by scoring psychological characteristics based on question items; and

[0060] An analyzing unit performs cluster analysis on the analysis data and extracts characteristic items from the psychological characteristic data for each cluster.

[0061] (Supplementary Note 2) An analysis system according to Supplementary Note 1, wherein:

[0062] further comprising an analysis data generating unit for generating the analysis data,

[0063] The storage unit stores a first database and a second database, wherein the first database is used to store the psychological characteristic data; and the second database is used to store purchase behavior data that determines purchase behavior based on numerical values.

[0064] The analytical data generating unit generates the analytical data by combining the purchase behavior data acquired from the second database with the psychological characteristic data acquired from the first database.

[0065] The analyzing unit extracts characteristic items from the purchase behavior data and the psychological characteristic data for each cluster.

[0066] (Supplementary Note 3) An analysis system according to Supplementary Note 1 or 2, wherein:

[0067] The psychological property is the interoceptive sensation.

[0068] (Supplementary Note 4) An analysis system according to Supplementary Note 2, wherein:

[0069] The device further includes a purchase quantification unit configured to collect the purchase behavior data during a predetermined period.

[0070] (Supplementary Note 5) An analysis system according to Supplementary Note 2 or 4, wherein:

[0071] The device further includes a drawing unit that generates graphic data for visually displaying an index of at least one of the purchase behavior data and the psychological characteristic data for each of the clusters.

[0072] (Supplementary Note 6) The analysis system according to any one of Supplementary Notes 1 to 5, wherein:

[0073] The psychological characteristic data includes scores of stress-related indicators obtained based on the questions.

[0074] (Supplementary Note 7) The analysis system according to any one of Supplementary Notes 1 to 6, wherein:

[0075] In the cluster analysis of the analytical data, the K-means method is used.

[0076] (Supplementary Note 8) An analysis system according to Supplementary Note 7, wherein:

[0077] The method further includes an auxiliary analysis unit configured to perform a principal component analysis when the number of clusters exceeds a predetermined number of clusters when cluster analysis is performed on the analysis data.

[0078] Description of Reference Numerals

[0079] 11: Analysis system (sensory behavior analysis system); 12: First database; 13: Storage unit (first storage unit); 14: Second database; 15: Storage unit (second storage unit); 17: Analysis data generation unit (data set generation unit); 18: Analysis unit; 22: Storage unit (third storage unit); 23: Auxiliary analysis unit; 24: Rendering unit; 27: Purchase digitization unit.

Claims

1. An analysis system comprising: a storage unit storing analysis data including psychological characteristic data obtained by scoring psychological characteristics based on question items; and An analyzing unit performs cluster analysis on the analysis data and extracts characteristic items from the psychological characteristic data for each cluster.

2. The analysis system according to claim 1, wherein: further comprising an analysis data generating unit for generating the analysis data, The storage unit stores a first database and a second database, the first database is used to store the psychological characteristic data, and the second database is used to store purchase behavior data that determines the purchase behavior based on a numerical value, The analytical data generating unit generates the analytical data by combining the purchase behavior data acquired from the second database with the psychological characteristic data acquired from the first database. The analyzing unit extracts characteristic items from the purchase behavior data and the psychological characteristic data for each cluster.

3. The analysis system according to claim 1 or 2, wherein: The psychological property is the interoceptive feeling.

4. The analysis system according to claim 2, wherein: The device further includes a purchase quantification unit configured to collect the purchase behavior data during a predetermined period.

5. The analysis system according to claim 2, wherein: The device further includes a drawing unit that generates graphic data for visually displaying an index of at least one of the purchase behavior data and the psychological characteristic data for each of the clusters.

6. The analysis system according to claim 1 or 2, wherein: The psychological characteristic data includes scores of stress-related indicators obtained based on the questions.

7. The analysis system according to claim 1 or 2, wherein: When performing cluster analysis on the analytical data, the K-means method is used.

8. The analysis system according to claim 7, wherein: The method further includes an auxiliary analysis unit configured to perform a principal component analysis when the number of clusters exceeds a predetermined number of clusters when cluster analysis is performed on the analysis data.

Images