Method for supporting evaluation and evaluation support system

Abstract

To construct a relation model which can identify how much related a feature element indicating the form, for example, of an evaluation target is to an intention of purchase.SOLUTION: The method for evaluation includes: a presentation step (S1) of presenting an evaluation target to a subject; an acquisition step (S2) of acquiring an answer of the subject to a question group of a plurality of questions each including at least one evaluation word related to an evaluation target; and a model creation step (S5) of creating a relation model including information on (1) plural features, (2) an impression, (3) an attitude, (4) an intention of purchase. The evaluation word includes words related to the impression, the attitude, and the intention of purchase.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an evaluation support method and an evaluation support system for creating an association model capable of identifying elements related to the purchase intention for an evaluation target.

Background Art

[0002] In Patent Document 1, a system is disclosed that creates a model obtained by performing multiple regression analysis based on the components (needs, wants, situation) of the customer's purchase will and customer attributes, and quantitatively analyzes the factors affecting the customer's purchase psychology.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, generally, the customer's purchase decision-making process is considered to go through a complex hierarchical structure, and it has been difficult to understand how elements indicating characteristics such as the form of a product and the intuitive mental movements (so-called sensibility) of consumers who have seen or experienced the product are linked to the formation of the consumer's attitude towards the product, and further how much they are related to the purchase intention, which is the upper layer of the consumer's attitude, by means of a simple two-layer regression model.

Means for Solving the Problems

[0005] In order to solve the above problems, an evaluation support method according to an aspect of the present invention includes a presentation step of presenting a subject with an evaluation target having a plurality of features, and for the evaluation target, a question including a question sentence related to each of the plurality of features and answer options for the question sentence, wherein at least one evaluation word related to the evaluation target is included in at least one of the question sentence and the answer options. An acquisition step of acquiring, by a computer, an answer of the subject to a plurality of question groups including the question; and a computer, based on each of the plurality of features, a word indicating a factor or component common to the evaluation words included in the question group, and the answer of the subject, (1) first information indicating each of the plurality of features, (2) second information regarding an impression received by the subject from the evaluation target based on each of the plurality of features, (3) third information regarding the attitude of the subject according to the impression, and (4) fourth information regarding the purchase intention of the subject according to the impression or the attitude. A model creation step of creating a correlation model including the above, wherein the evaluation words include words related to the impression, words related to the attitude, and words related to the purchase intention.

[0006] In order to solve the above problems, an evaluation support system according to an aspect of the present invention includes, for an evaluation target having a plurality of characteristics presented to a subject, a question including a question sentence related to each of the plurality of characteristics and answer options for the question sentence, wherein at least one of the question sentence and the answer options includes at least one evaluation word related to the evaluation target, an acquisition unit that acquires the subject's answers to a plurality of questions included in a question group, and, based on each of the plurality of characteristics, a word indicating a factor or component common to the evaluation words included in the question group, and the subject's answers, (1) first information indicating each of the plurality of characteristics, (2) second information regarding the impression received by the subject from the evaluation target based on each of the plurality of characteristics, (3) third information regarding the subject's attitude according to the impression, and (4) fourth information regarding the subject's purchase intention according to the impression or the attitude, a model creation unit that creates a correlation model including the above, and the evaluation words include words related to the impression, words related to the attitude, and words related to the purchase intention.

[0007] An evaluation support system capable of executing an evaluation support method according to each aspect of the present invention may be realized by a computer. In this case, an evaluation support system control program that causes the computer to operate as each part (software element) included in the evaluation support system to realize the evaluation support system by the computer, and a computer-readable recording medium on which it is recorded also fall within the scope of the present invention.

Effects of the Invention

[0008] According to an aspect of the present invention, it is possible to analyze, using a correlation model, how much the elements indicating the characteristics such as the form of the evaluation target are related to the purchase intention.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Mode for Carrying Out the Invention

[0010] 〔Embodiment 1〕 Hereinafter, an embodiment of the present invention will be described in detail.

[0011] <Overview of the Evaluation Support Method> In this embodiment, a method for creating an association model for analyzing elements related to the purchase intention of an evaluation target and the degree of association between the purchase intention and each element using a question group will be described.

[0012] The evaluation target is an object to be evaluated by a subject. The evaluation target may be the object itself such as a product, or an experience related to the object to be evaluated (for example, an image, description of a certain product, or service received in a certain environment). A subject is a person who is presented with an evaluation target and answers questions regarding the evaluation target.

[0013] In this specification, an element refers to an element considered to be related to the purchase intention of a consumer towards an evaluation target, such as an impression on the evaluation target including the subject, and attitudes such as the subject's perception, emotion, and behavioral tendency towards the evaluation target, and is expressed by words indicating the element. An element indicates the subject's response to a group of questions including evaluation words related to the evaluation target, or a factor or component derived by statistically analyzing the response, and may be represented by words representing the layers of the correlation model or words included in each layer. The elements used in the method according to the present disclosure may include characteristic elements, impression elements, attitude elements, and purchase intention elements. The elements used in this specification are numerical values or words, and are labeled by the words representing each of the elements.

[0014] An impression element is an element indicating the impression that a subject has towards an evaluation target. The impression element may include an element indicating an instinctive impression held from the appearance and touch of the evaluation target, etc., and an element indicating an impression on functions such as usability and performance when imagined from the appearance of the evaluation target or actually used. The impression element may further include an element indicating an introspective impression associated with an individual's memory and association. Furthermore, the impression element may be classified into an element indicating an impression relatively closer to the subject's emotion (higher-order impression) and an element indicating an impression relatively closer to the morphological element of the evaluation target (lower-order impression). An attitude element is an element indicating what attitude a subject takes towards an evaluation target. The attitude element may be further classified into a cognitive element, an emotional element, and a behavioral element. The cognitive element is an element indicating the subject's perception of the evaluation target, the emotional element is an element indicating the emotion that the subject has towards the evaluation target, and the behavioral element is an element indicating the behavioral tendency of the subject towards the evaluation target. Also, the purchase intention element is an element indicating the purchase intention itself, indicating whether the subject decides to purchase the evaluation target.

[0015] Furthermore, the elements used in the method according to the present disclosure may further include other elements. For example, the elements may include elements indicating economic situation, recommendations from others, the intensity of interest in the product, the confidence in being able to master the product, etc. These may also be expressed as elements indicating "subjective norm" or "sense of behavioral control" proposed in the theory of planned behavior in consumer behavior research.

[0016] FIG. 1 is a flowchart for explaining an example of the flow of the evaluation support method according to Embodiment 1 of the present disclosure. As shown in FIG. 1, the evaluation support method according to the present disclosure includes a presentation step (S1), an acquisition step (S2), and a model creation step (S5). The evaluation support method may be performed using a computer such as a general-purpose computer.

[0017] The presentation step is a step of presenting an evaluation target having a plurality of features to a subject. The acquisition step is a step of acquiring responses to the results of presenting a group of questions related to the evaluation target to a plurality of subjects. Note that the subject may be a person who is in daily contact with the evaluation target. Alternatively, the subject may be a person who wants to be in contact with the evaluation target or the like.

[0018] The plurality of features of the evaluation target are features related to the five senses and other senses, and may be either a measurement result related to the nature of the evaluation target or an evaluation result obtained by evaluating the features based on a predetermined criterion. Each feature is represented by a word corresponding to each of the features. For example, the features of the evaluation target may be morphological characteristics of the evaluation target that can be evaluated by measuring the evaluation target, such as the shape, size, weight, color, temperature, friction, or volume of the evaluation target. Further, the features of the evaluation target may be an evaluation result indicating which of a plurality of pre-divided stages or presence / absence the characteristics of what the evaluation target corresponds to. Note that the features are not limited to the morphological characteristics of the evaluation target. For example, the features may be the sound emitted from the evaluation target, the taste of the evaluation target, the smell of the evaluation target, and the information recorded on the evaluation target, etc.

[0019] The question includes at least one evaluation word related to the evaluation target in at least one of the question text and the answer options. The evaluation word is a word used to evaluate the evaluation target, and words related to the characteristics, impressions, attitudes, and purchase intentions when the evaluator contacts the evaluation target are pre-selected according to the evaluation target and pre-classified according to the nature of the words. The evaluation word may include words representing elements related to the purchase intention of the subject. For example, the evaluation word includes words indicating the characteristics of the evaluation target, words related to the impression of the evaluation target, words related to the attitude towards the evaluation target, and words related to the purchase intention towards the evaluation target.

[0020] The extraction process is a process of extracting elements used in the association model to be created in the subsequent model creation process from the answers obtained in the acquisition process. In the extraction process, factors or components common to the evaluation words included in the question group are extracted layer by layer. A layer is a plurality of layers that make up the association model created in the model creation process, and each layer and element are pre-classified according to the nature of the words for classifying the evaluation words according to the words indicating each layer and element.

[0021] The model creation process is a process of creating an association model using the answers obtained in the acquisition process and the elements extracted based on the answers, and obtaining information indicating the relationship between the elements.

[0022] FIG. 2 is a diagram showing an overview of an association model created by the method according to Embodiment 1 of the present disclosure. The association model created in the model creation step includes elements of a sensory value model and an effect hierarchy model (ABC model) based on three elements of attitude. These three elements may be arranged in any order in different hierarchies within the layer indicating the attitude element, or any plurality of elements may be arranged in the same hierarchy. Furthermore, a hierarchical structure may be assumed within one element. Also, the impression element may be divided into an element indicating a lower-order impression and an element indicating a higher-order impression, and may show a hierarchical structure from the lower-order impression to the higher-order impression. Also, the association model may include elements that occur later in time series than the purchase intention, such as purchase behavior, consumption, post-purchase evaluation, and disposal, as a higher hierarchy than the purchase intention.

[0023] As shown in FIG. 2, the created association model includes first information, second information, third information, and fourth information shown in the following (1) to (4). The first information is related to the characteristics of the evaluation target, the second information is related to the impression element, the third information is related to the attitude element, and the fourth information is related to the purchase intention element.

[0024] (1) First information: Information indicating each of a plurality of characteristics of the evaluation target (reference numeral 202 in FIG. 2) (2) Second information: Information regarding the impression received by the subject from the evaluation target based on each of the plurality of characteristics (reference numeral 203 in FIG. 2) (3) Third information: Information regarding the attitude of the subject towards the evaluation target according to the impression (reference numeral 204 in FIG. 2) (4) Fourth information: Information regarding the purchase intention of the subject towards the evaluation target according to the impression or attitude (reference numeral 205 in FIG. 2)

[0025] Also, the association model may include fifth information, which is other information other than the first to fourth information, such as subjective norms, sense of behavioral control, or the environment / situation in which the subject is placed, which are defined independently of the subject's sensibility. The environment / situation in which the subject is placed may be, for example, the subject's prior knowledge of the evaluation target, the subject's interest in the evaluation target, or the subject's purchasing power.

[0026] Also, in the association model, the third information may be further classified by a plurality of more detailed elements. Specifically, the third information may be classified into a cognitive element (reference numeral 208 in FIG. 2), an emotional element (reference numeral 209 in FIG. 2), and an action element (reference numeral 210 in FIG. 2). Note that, in the association model, the hierarchical structure of the three elements included in the third information is not limited to that shown in FIG. 2, and may be changed according to the evaluation target and the attributes of the subject.

[0027] Furthermore, the association model includes information indicating the degree of association from the elements included in the lower hierarchy to the elements included in the upper hierarchy.

[0028] Specifically, the association model includes information indicating the degree of association between at least any one of the impression element included in the second information, the attitude element included in the third information, and the purchase intention element included in the fourth information, based on the characteristic elements of the evaluation target included in the first information. The association model also includes information indicating the degree of association between at least any one of the attitude element included in the third information and the purchase intention element included in the fourth information, based on the impression element included in the second information. Furthermore, the association model includes information indicating the degree of association between the attitude element included in the third information and the purchase intention element included in the fourth information. Also, when the association model includes fifth information, the association model may include information indicating the degree of association between at least any one of the purchase intention element, the attitude element, and the impression element, based on the element included in the fifth information.

[0029] Conventionally, in the sensory value model, the relationship from the characteristics of the evaluation target to human emotions could be specified, but it was unclear how these characteristics and emotions were related to the purchase intention. For example, it often occurred that even though a consumer had a favorable impression of the evaluation target, the consumer did not actually make a purchase.

[0030] In addition, in the ABC model of attitudes, while it was possible to identify what kind of attitudes of people are linked to purchase intentions, it was unclear what kind of characteristic elements or impression elements of the evaluation target are related to consumers' attitudes and how. Therefore, it was not possible to analyze for any product what kind of product can improve people's purchase intentions.

[0031] On the other hand, the association model created by the evaluation support method according to the present disclosure includes a sensory value model (reference numeral 206 in FIG. 2), an ABC model of attitudes (reference numeral 207 in FIG. 2), and elements of purchase intention, and further includes information indicating the relationships between the elements included in these models.

[0032] Therefore, the association model created by the evaluation support method according to the present disclosure can analyze how the characteristics of the evaluation target or the impressions / attitudes of the subjects who are consumers towards the evaluation target are related to the purchase intention towards the evaluation target. Based on the association model created by the method according to the present disclosure, a product designer can analyze what kind of characteristics of the evaluation target can improve or decrease the purchase intention towards the evaluation target. Also, the product designer can understand the characteristics that improve consumers' purchase intention based on the association model and utilize them in product design. Therefore, the evaluation support method according to the present disclosure can also be applied to product design methods.

[0033] <Evaluation Support System 100> The evaluation support method according to this embodiment may be performed by a system capable of executing each step of the method. The evaluation support method according to this embodiment is executed by one or more computers (including tablet terminals and smartphones) capable of executing each step of the method. FIG. 3 is a block diagram showing an example of the configuration of an evaluation support system 100 capable of executing the evaluation support method according to Embodiment 1. As shown in FIG. 3, the evaluation support system 100 includes an input device 1 and a model creation device 2. Note that the input device 1 and the model creation device 2 may be separate computers or may be one computer. For example, one computer may have the functions of the input device 1 and the model creation device 2. Furthermore, each part within the model creation device 2 may be a separate computer, or the input device 1 and the display unit 22 may be the same computer. The input device 1 is a device for inputting information indicating an answer to a question by a subject. The answer may be input using the input device 1 by the subject who was asked the question or by someone who heard the subject's answer.

[0034] As shown in FIG. 3, the model creation device 2 includes a control unit 21 and a display unit 22. The model creation device 2 may also include a storage unit 23 that stores various information used in the model creation device 2, such as information indicating a question group, information indicating an input answer, and information indicating an analysis result based on the answer. Answers obtained by presenting a question group are input to the model creation device 2. The answer may be input using the input device 1 such as a keyboard, for example.

[0035] The control unit 21 includes an acquisition unit 211, an analysis unit 212, and a model creation unit 213. The acquisition unit 211 acquires information indicating an answer to a question group. The analysis unit 212 performs analysis based on the acquired answer and extracts the load or score of factors or components representing the evaluation words included in each layer, or the elements representing the evaluation words included in each layer. Details of the analysis performed by the analysis unit 212 will be described later.

[0036] Based on the analysis results by the analysis unit 212, the model creation unit 213 selects the components of the association model and creates the association model. Details of the method for creating the association model will be described later. The model creation unit 213 causes the created association model to be displayed on the display unit 22. Note that the model creation unit 213 may transmit the association model to another device communicably connected to the model creation device 2.

[0037] The association model created by the model creation unit 213 includes the first to fourth information, and in some cases, the fifth information. The association model also includes information indicating the degree of association from the elements included in one layer included in the association model to the elements included in the upper layer. For example, the association model includes information indicating the degree of association from the characteristic elements of the evaluation target to the impression elements, the degree of association from the impression elements to the attitude elements, and the degree of association from the attitude elements to the purchase intention elements.

[0038] As described above, the model creation device 2 can create an association model capable of analyzing the degree of association between elements such as characteristics in the evaluation target based on the answers from the subjects to the question group regarding the evaluation target. Those who have confirmed the association model, such as product designers, can analyze how each characteristic of the evaluation target is related to the purchase intention of consumers.

[0039] <Evaluation Support Method> Hereinafter, the details of the evaluation support method will be described again with reference to FIG. 1. As an example, the evaluation support method is performed using the model creation device 2. Note that a part of the evaluation support method may be performed by a person. For example, in the evaluation support method, the model creation process may be executed by a person who has obtained the analysis result in the acquisition process and understands it.

[0040] First, in the evaluation support method, the subject to be evaluated is presented with the object to be evaluated (S1: presentation step). In the presentation step, the object to be evaluated is presented while assuming a situation where the subject considers purchasing the object to be evaluated. In addition, in the presentation step, in addition to the purchase of the object to be evaluated, other situations related to the object to be evaluated may also be assumed. For example, in the presentation step, the subject may be made to assume a situation of "moving to and renovating an old house, setting a budget of X yen in advance, and then selecting the floor material for one's own study", and then presenting the wood pieces that can be used as the floor material as the object to be evaluated.

[0041] "Presenting the object to be evaluated" may mean showing the actual object of the object to be evaluated, or may mean giving the subject an experience related to the object to be evaluated. For example, it may mean showing the subject an image of the object to be evaluated, making the subject hear a sound, smell a smell, taste the object to be evaluated, touch the object to be evaluated, or presenting the information recorded on the object to be evaluated, etc., that is, giving the subject an experience including these. Specifically, if the object to be evaluated is wood, the subject may be allowed to touch the actual wood. Also, if the object to be evaluated is a music CD, in the presentation step, the subject may be shown the appearance of the object to be evaluated and allowed to audition the music recorded on the CD.

[0042] Note that the number of objects to be evaluated presented to the subject may be one or more. For example, the subject may be made to assume a situation of comparing and considering the objects to be evaluated, and a plurality of comparable objects to be evaluated may be presented. Also, if there is one article presented to the subject, a plurality of conditions related to the article may be set, and each of these plurality of conditions may be used as a plurality of objects to be evaluated. The presentation order of these objects to be evaluated may be different (random) for each subject, or may be in the same order.

[0043] After prompting engineering, a group of questions related to the evaluation target is presented to the subject to obtain answers (S2). The group of questions includes a plurality of sets of questions each containing a question sentence related to each of a plurality of features and answer options for the question sentence. Further, the question includes at least one evaluation word related to the evaluation target in at least one of the question sentence and the answer options. Note that the method of presenting the group of questions is not particularly limited. The group of questions may be displayed, for example, on a presentation unit such as a display, printed on paper, or conveyed orally to the subject.

[0044] The evaluation words included in the group of questions include (1) words related to the impression the subject receives from the evaluation target, (2) words related to the subject's attitude according to the impression, and (3) words indicating the subject's purchase intention according to the impression or attitude. Further, the evaluation words included in the group of questions may include words indicating subjective norms, words indicating a sense of action control, and words indicating other external elements, or may include words indicating elements after the purchase intention (purchase behavior, consumption, post-purchase evaluation, disposal, etc.).

[0045] The collection of the evaluation words may be carried out through a pre-interview with the subject, may be borrowed from related existing literature, or may be determined through consultation among related parties. The interview may be conducted by any method or may follow existing methods such as the evaluation grid method.

[0046] Words related to the impression may include, for example, words related to higher-order impressions such as "simple", "elegant", "affectionate", "seemingly easy to get dirty", and words related to lower-order impressions such as "coarse", "smooth", "seemingly not easily damaged".

[0047] Words related to attitude may include words indicating the subject's cognition, words indicating emotion, and words indicating behavioral tendencies. Words indicating cognition may be words such as "good", "bad", "matching the image", and "worthwhile". Words indicating emotion may be words such as "like" and "dislike", "pleasant", "unpleasant", "awaken", and "calm down". Words indicating behavioral tendencies may be words such as "want" and "don't want", "want to buy", "want to go".

[0048] Words indicating purchase intention may be words such as "buy" and "not buy".

[0049] Words indicating subjective norms may be words such as "care about others' opinions". Also, words indicating a sense of behavioral control may be words such as "seem to be able to master (the product)". Also, words indicating other external factors may be words such as "can buy", "have knowledge", "have interest".

[0050] The method of answering questions may be a method in which the subject answers by selecting a numerical value in an arbitrary range such as 1 to 5 for the relevance of the evaluation words for the evaluation object. For example, when a question such as "Do you think the evaluation object is good?" is presented as a question regarding the subject's cognition, numbers from 1 to 5 may be presented as answer options. The subject may answer by selecting 1 if they feel "bad" and 5 if they feel "good". Also, when the subject feels that the evaluation object is at an intermediate level between "good" and "bad", they may answer by selecting 2 to 4 according to the degree of goodness. Note that it is not necessary to necessarily correspond the magnitude of the numerical value with the goodness or badness of the evaluation. The subject may be made to select 5 if they feel good and 1 if they feel bad, or the subject may be made to select 5 if they feel bad and 1 if they feel good. Also, the question form may present two opposite words as above, or may present one word and ask the subject to answer numerically the degree of match between that word and the evaluation object.

[0051] Alternatively, the method of answering questions may be that the subject answers by selecting one from a plurality of words for a question sentence including one evaluation word for the evaluation target. For example, as a question regarding the subject's cognition, when a question "Do you think the evaluation target is good?" is presented, words such as "bad", "slightly bad", "ordinary", "slightly good", and "good" may be presented as answer options. The subject may answer by selecting any of the options based on their own sensibility.

[0052] When a plurality of evaluation targets are presented to the subject, the subject answers for each of the plurality of evaluation targets.

[0053] The acquisition unit 211 acquires the answers to the results of presenting a question group to a plurality of subjects (S2: acquisition step). Note that in one creation of the association model, the presentation step and the acquisition step may be repeated a plurality of times. When the presentation step and the acquisition step are performed a plurality of times, the question group and the answer options presented in the presentation step may be common or different.

[0054] The acquisition unit 211 identifies the evaluation words included in the question group, classifies the evaluation words for each layer, and outputs information indicating the classification result and information indicating the answers to the analysis unit 212. Here, the layer indicates the nature of words such as attitude (behavior tendency, emotion, cognition), impression (higher order), and impression (lower order). Classifying the evaluation words for each layer means classifying the evaluation words according to the nature of the words. The layer is used to represent each element in a hierarchical structure in the association model described later. The layer is divided into, for example, four layers: a layer including first information (feature element), a layer including second information (impression element), a layer including third information (attitude element), and a layer including fourth information (purchase intention element). Further, the layer including the second information may be divided into two layers: a layer including higher order impressions and a layer including lower order impressions. Further, the layer including the third information may be divided into three layers: a layer including a "cognition element", a layer including an "emotion element", and a layer including an "action (tendency) element".

[0055] It may be determined in advance which element each evaluation word corresponds to and which layer it is classified into. The acquisition unit 211 may output the pre-classified evaluation words to the analysis unit 212 together with the response information.

[0056] When the analysis unit 212 acquires information indicating evaluation words classified by layer, it extracts factors or components common to the evaluation words included in one layer (S3: extraction step). The extraction of factors or components is performed using, for example, statistical analysis. The analysis unit 212 performs such extraction for each of the respective layers. The word indicating the factor or component may be any of the evaluation words included in the layer, or may be another word representing the evaluation words included in the layer. Also, one factor or component may be extracted for one layer, or a plurality of factors or components may be extracted for one layer. Note that the extraction step is performed only when there are many acquired evaluation words, and may be omitted when there are few evaluation words. The determination of whether there are many or few evaluation words may be appropriately determined by the analyst involved in creating the association model. For example, when it is determined by the analyst that there are many evaluation words, the analysis unit 212 receives an instruction from the analyst and extracts factors or components common to the evaluation words included in one layer, and when it is determined that there are few evaluation words, the extraction step may be omitted.

[0057] When the evaluation words included in the question group are used as they are in creating the association model, the number of words (elements) may be too large, making it difficult to analyze the relationship between the upper and lower hierarchies. By the analysis unit 212 extracting factors or components from the evaluation words, the number of elements used in creating the association model can be reduced. By using factors or components obtained by aggregating the evaluation words, it becomes possible to create an association model in which the relationship between the elements is easier to analyze. Here, a factor is intended to be an element that affects the events and concepts represented by the evaluation words and is extracted by factor analysis. Also, a component is intended to be an element represented by a set of events and concepts represented by the evaluation words and is integrated by principal component analysis.

[0058] In particular, for evaluation words that tend to be numerous, a small number of latent factors may be extracted by factor analysis, quantification theory type III, etc. By doing so, a more reliable model can be constructed by eliminating the similarity between evaluation words. For example, when factor analysis is performed, the maximum likelihood method, etc. can be adopted for the factor extraction method, and promax rotation, etc. can be adopted for factor rotation.

[0059] The analysis unit 212 performs analysis using information indicating factors or components, information indicating the responses of the subjects, and information indicating the characteristics of the evaluation target, and quantitatively determines the loading that represents how much influence the evaluation words included in the layer have on the factor or component representing the layer, and the score that represents how much influence the evaluation target has on the factor or component (S4). The analysis unit 212 outputs information indicating the analysis result to the model creation unit 213.

[0060] For example, the analysis unit 212 performs factor analysis on the response data obtained from the subjects, extracts the factor loadings and factor scores of the factors representing the evaluation words included in each layer, and performs a process of representing the obtained response data matrix as the product of the loading matrix and the score matrix.

[0061] Also, for example, the analysis unit 212 may perform principal component analysis instead of factor analysis to determine the principal component loadings and principal component scores, or may adopt a method such as quantification theory type III according to the nature (scale level) of the element data. Alternatively, as these alternative methods, various cluster analyses (k-means method, hierarchical clustering) or multidimensional scaling methods may be used to group a plurality of elements into a small number of representative elements.

[0062] When the model creation unit 213 obtains the analysis result by the analysis unit 212, it creates an association model based on the analysis result (S5: model creation step). The analysis result by the analysis unit 212 includes, for example, a loading (loading) indicating the degree of association between an evaluation word and a common factor or a principal component obtained based on an answer obtained from a subject, a score (score) indicating the degree of association between an evaluation target and a common factor or a principal component, or a representative value of answers to a plurality of questions with similar answer tendencies of the subject. Note that appropriate words may be assigned to the common factor, the principal component, or the representative value by an analyst. The model creation unit 213 obtains information indicating words corresponding to the common factor, the principal component, or the representative value, for example, from the input device 1 operated by the analyst.

[0063] The model creation step can also be expressed as a step of creating an association model including information indicating the relationship between the elements (common factor, principal component, or representative value) extracted for each layer, represented by words, based on the extracted elements (common factor, principal component, or representative value) and the answers from the subject. The association model created by the model creation unit 213 is a hierarchical model created using any of statistical modeling methods such as multiple regression, covariance structure analysis, Bayesian network, or other modeling methods such as rough set and network analysis.

[0064] For example, the model creation unit 213 performs, for example, covariance structure analysis (SEM) using the scores (scores) of the factors or components identified by the analysis unit 212 or the representative values of a plurality of answers. Thereby, the model creation unit 213 determines a value indicating the degree of association between each element and the upper-level element.

[0065] For example, based on the analysis results by the analysis unit 212, the model creation unit 213 identifies the degree of association between each element included in the layer indicating features and the elements included in the layer indicating impressions. Similarly, the model creation unit 213 identifies the degree of association between impression elements and attitudes, and between attitude elements and purchase intention elements. Note that the association of one element with another element performed by the model creation unit 213 is not limited to the above-described ones. For example, the model creation unit 213 may identify the degree of association between each element included in the layer indicating features and the elements included in the layer indicating impressions, or further the three elements of attitude and purchase intention elements located in the upper layer. Alternatively, when impression elements are divided into lower-order impression elements and higher-order impression elements, the degree of association between lower-order impression elements and higher-order impression elements, or the degree of association between lower-order / higher-order impression elements and the three elements of attitude and purchase intention elements may be identified.

[0066] Based on the above specific results, the model creation unit 213 creates an association model including information indicating the relationship between a plurality of elements. The association model created in this way includes the following information (1) to (4). (1) First information indicating each of a plurality of features (2) Second information regarding the impressions received by the subject from the evaluation target based on each of the plurality of features (3) Third information regarding the attitude of the subject according to the impression (4) Fourth information regarding the purchase intention of the subject according to the impression or attitude

[0067] The second information may include information indicating lower-order impression elements and higher-order impression elements. Further, the third information includes information indicating the cognition, emotion, and behavioral tendency of the subject with respect to the evaluation target.

[0068] Furthermore, the association model includes information indicating the degree of association between elements included in the lower layer and elements included in the upper layer. For example, in the association model, the first information includes information indicating the degree of association between each of a plurality of feature elements and the impression element. Also, the second information includes information indicating the degree of association between each of the impression elements and each of the three elements of attitude. Further, the third information includes information indicating the degree of association between each of the three elements of attitude and the purchase intention element. Note that the first to third information may further include information indicating the degree of association with other elements. For example, the first information may include information indicating the degree of association between each of the plurality of feature elements and the three elements of attitude or the purchase intention element.

[0069] Also, when the association model includes fifth information, the fifth information may include information indicating the degree of association between elements other than the first to fourth information and the purchase intention element, the attitude element, or the impression element. For example, the fifth information may include information indicating the degree of association between an element indicating the economic situation and financial resources of the subject and the purchase intention element.

[0070] Further, the model creation unit 213 may confirm the validity of the model based on several fitness indicators obtained as a result of the statistical modeling method. In this case, when using a regression statistical method, the model creation unit 213 may specify the coefficient of determination of each element, and when using covariance structure analysis, may also specify the values of GFI (AGFI) and RMSEA in addition thereto. The model creation unit 213 may adopt, as the association model regarding the evaluation target, a model in which GFI (AGFI) is as large as possible in the range of 0 to 1 and RMSEA is as small as possible.

[0071] After the model creation process, the model creation unit 213 outputs the created association model (S6). By checking the output association model, a product designer or the like can analyze which features of the evaluation target evoke which emotions of consumers, and which features lead to a purchase intention, and so on.

[0072] <Specific Example of Association Model> FIG. 4 is a diagram showing a specific example of a correlation model created using the evaluation support method according to the present disclosure. The model creation unit 213 may create and output a diagram (reference numeral 401 in FIG. 4) showing a correlation model as shown in FIG. 4. Hereinafter, a specific example of a method for creating a correlation model will be described with reference to FIG. 4.

[0073] In the evaluation support method, first, a presentation step is performed. Specifically, a situation of considering the purchase of a floor material for a study is presented to the subject, along with physical samples of pieces of wood of a plurality of types as candidates for the floor material and the price per unit quantity of each type of wood. After the presentation step, a question group is presented to the subject to obtain answers. The evaluation words included in the questions in the question group or the answer options are shown in Table 1.

[0074] [Table 1]

[0075] As shown in Table 1, the question group includes at least one evaluation word each indicating impression, recognition, emotion, behavior, and purchase intention. Also, as shown in Table 1, the evaluation words indicating impression include words of instinctive items related to the appearance and touch of the evaluation object, etc., words of functional items related to the usability and performance of the evaluation object, etc., and words of introspective items related to the personal interpretation and inference of the subject. Note that the words of the introspective items may be included in the cognitive element, which is one level higher, depending on the judgment of the analyst.

[0076] Regarding the scale pair of "buy / not buy" which is the purchase intention, for example, after all other questions regarding all types of wood have been completed, three options of "not buy", "consider", and "buy" may be presented to the subject for each type of wood to obtain answers. Note that the option of "buy" may be selectable for at most one type of wood, and the options of "consider" and "not buy" may be selectable for any number of types of wood.

[0077] The acquisition unit 211 acquires, for example, the answers input by the questioner or the subject using the input device 1 (acquisition step), and outputs information indicating the acquired answers and information indicating the evaluation words included in the question group to each part of the control unit 21.

[0078] The analysis unit 212 extracts factors or components based on the answers to the question group including evaluation words. For example, the analysis unit 212 extracts the factors or components common to the evaluation words included in the question group for each layer that classifies the nature of the words. For example, the analysis unit 212 extracts at least one feature element, impression element, cognitive element, emotional element, behavioral element, and purchase intention element respectively based on the characteristics of the evaluation target and the evaluation words. As an example, the number of factors extracted by the analysis unit 212 from the impression element by factor analysis may be 3 factors related to instinctive items such as power (thick and firm), activity (full of naturalness), and evaluativeness (friendly), and 1 factor related to functional items such as durability (not easily damaged and not easily soiled). In this specific example, the analysis unit 212 performs factor extraction, but the step in which the analysis unit 212 performs factor or component extraction is not essential and may be omitted.

[0079] In addition, the analysis unit 212 specifies the evaluation score of each question included in the question group. For example, when the subject selects the answer "like" for the question "Do you like the evaluation target?", the analysis unit 212 sets the evaluation score of the question to 5 points, and when the answer "dislike" is selected, it sets it to 1 point. For questions regarding purchase intention, the analysis unit 212 may also assign evaluation scores with "buy" as 2, "consider" as 1, and "not buy" as 0.

[0080] The analysis unit 212 performs statistical analysis such as factor analysis and principal component analysis based on information indicating the evaluation points of the subject's answers and information indicating the characteristics of the evaluation target, and calculates the loadings, scores, or representative values corresponding to each element. Examples of the analysis results by the analysis unit 212 are shown in Table 2 and Table 3. Table 2 shows the degree of association (factor loading) between each factor included in the instinctive items among the impression elements and the evaluation words, and the factor contribution of each factor. Table 3 shows the degree of association (factor loading) between each factor included in the functional items among the impression elements and the evaluation words, and the factor contribution of each factor.

[0081]

Table 2

[0082]

Table 3

[0083] Based on the analysis results of the analysis unit 212, the model creation unit 213 performs a statistical modeling method or other modeling method to determine a value indicating the degree of association between each element and the upper-level element. The model creation unit 213 creates an association model 401 based on the analysis results. FIG. 4 is a diagram showing an example of the association model 401 created by the model creation unit 213.

[0084] As shown in FIG. 4, the association model 401 has a layer including feature elements (reference numeral 402), a layer including impression elements (reference numeral 403), a layer including attitude elements (reference numeral 404), a layer including purchase intention elements (reference numeral 408), and a layer including other elements (reference numeral 409). Further, the layer including attitude elements includes a layer including cognitive elements (reference numeral 405), a layer including emotional elements (reference numeral 406), and a layer including behavioral elements (reference numeral 407). Note that the hierarchical structure shown in FIG. 4 is an example, and other hierarchical structures may be adopted. For example, in the association model, the hierarchical structure of cognitive elements, emotional elements, and behavioral elements is not limited to the structure shown in FIG. 4.

[0085] In the association model shown in FIG. 4, with the layer containing the feature elements as the lowest layer and the layer containing the purchase intention elements as the highest layer, it shows how the lower elements are related to the higher elements. The numerical values near the arrows indicate the degree of association from each element to another element. It can be said that the stronger the relationship from the element at the origin of the arrow to the element at the tip of the arrow, the larger the absolute value of the described numerical value.

[0086] The arrows shown by solid lines in FIG. 4 indicate a positive correlation between elements. In the association model 401, when the lower element increases, the related higher element also increases. For example, there is a positive correlation between the element "glossiness" and the element "evaluability". In this case, it shows that the higher the glossiness of the evaluation target, the greater the impression regarding "evaluability" such as "being fond of" that the subject has towards the evaluation target.

[0087] Also, the arrows shown by dashed lines in FIG. 4 indicate a negative correlation between elements. In the association model 401, when the lower element increases, the related higher element decreases. For example, there is a negative correlation between the element "brightness" and the element "durability (·blackness)". In this case, it shows that the higher the brightness of the evaluation target, the smaller the impression regarding "durability (·blackness)" such as "the dirt is not noticeable" that the subject has towards the evaluation target.

[0088] Also, in FIG. 4, for the feature elements "blue, yellow", it shows that the value becomes larger as the evaluation target is closer to yellow than blue, and the value becomes smaller as it is closer to blue than yellow. For example, as shown in FIG. 4, there is a negative correlation between the feature element "blue, yellow" and the impression element "good association". Here, it can be seen that when the evaluation target has a color closer to yellow than blue, it has a negative impact on "good association".

[0089] Also, the model creation unit 213 may calculate the relative contribution rate of each of the feature elements to the purchase intention element and output the calculation result. Table 4 shows an example of the relative contribution rate of each of the feature elements to the purchase intention element.

[0090]

Table 4

[0091] In the example shown in FIG. 4, it can be seen that for the purchase intention of the subject, a cognitive element of "image match" and a behavioral element of "want" are related. Also, in the example shown in FIG. 4, it can be seen that the cognitive element that the evaluation target matches the image is more strongly related than the behavioral element of "want to have".

[0092] Also, in the example shown in FIG. 4, it can be seen that the impression element of "durability (·blackness)" is most strongly related to "image match", and the characteristic elements of "lightness", "dynamic friction coefficient", and "specific gravity" are related to "durability (·blackness)". Furthermore, it can be seen that "lightness" is most strongly related to "durability (·blackness)".

[0093] As described above, in the situation of "selecting a floor material for a study", it can be seen that for the subject, "matching the image" is more strongly related to the purchase intention than "wanting" the evaluation target. Furthermore, it can be seen that the lightness of the wood as the evaluation target is most strongly related to "matching the image", and furthermore, it has the most positive relationship with the cognition that "matching the image" is achieved by the low lightness.

[0094] On the other hand, in the example shown in FIG. 4, when the wood as the evaluation target is closer to red than to green, a good association is held for the wood, leading to the cognition of good and the emotion of liking. However, the relationship between the emotion of liking and the purchase intention through "want" is weaker than the relationship between the image match and the purchase intention. Therefore, it can be seen that although the subject has a good impression of the wood closer to red than to green, the degree of association with the improvement of the purchase intention is not so large.

[0095] As shown in FIG. 4 and Table 4, in this specific example, the relationship with the purchase intention is strongest for lightness, and then color tone and dynamic friction coefficient are related. Considering the signs of the coefficients of each element, it can be interpreted that the wood of "dark black with little friction" evoked the purchase desire of the experimental participants as the floor material for the study.

[0096] 〔Embodiment 2〕 Another embodiment of the present invention will be described below. For convenience of explanation, members having the same functions as those described in the above embodiment are denoted by the same reference numerals, and the description thereof will not be repeated.

[0097] The evaluation support method according to Embodiment 2 includes a first prediction step of calculating a predicted value of information that has not been newly input among the first information and the fourth information from the newly input first information or the newly input fourth information based on the newly input first information or the newly input fourth information to the correlation model.

[0098] FIG. 5 is a flowchart showing an example of the flow of the evaluation support method according to Embodiment 2. As shown in FIG. 5, the evaluation support method according to Embodiment 2 includes a first prediction step (S18) of calculating a predicted value of the fourth information based on the first information, and a first prediction step (S20) of calculating a predicted value of the first information based on the fourth information. Note that the evaluation support method according to Embodiment 2 may include only one of the first prediction step (S18) of calculating a predicted value of the fourth information based on the first information and the first prediction step (S20) of calculating a predicted value of the first information based on the fourth information.

[0099] FIG. 6 is a block diagram showing the configuration of an evaluation support system 100A according to Embodiment 2. As shown in FIG. 6, the evaluation support system 100A includes an input device 1 and a model creation device 2A. The model creation device 2A includes a control unit 21A having a prediction unit 214.

[0100] Hereinafter, the flow of the evaluation support method according to Embodiment 2 will be described with reference to FIGS. 5 and 6. Note that the processes of steps S11 to S16 in FIG. 5 are the same as the processes of steps S1 to S6 described with reference to FIG. 1 in Embodiment 1, and thus the description thereof will be omitted.

[0101] After the association model is created by the model creation unit 213 and displayed on the display unit 22, the prediction unit 214 determines whether new first information has been input (S17). The input of the new first information is performed, for example, by an analyst who performs analysis using the association model using the input device 1. The new first information may be a numerical value that changes the value of the feature element included in the association model.

[0102] When new first information is input (YES in S17), the prediction unit 214 calculates a predicted value of the fourth information (purchase intention) when the new first information is input to the association model (S18: first prediction step). For example, the prediction unit 214 predicts the value of the fourth information after the input of the new first information. The prediction unit 214 causes the display unit 22 to present a prediction result indicating the predicted relevance (S21). Note that the prediction unit 214 may calculate a predicted value of the second information based on the new first information, and further calculate a predicted value of the third information using the new first information and the second information calculated based on the new first information.

[0103] When no new first information has been input (NO in S17), the prediction unit 214 determines whether new fourth information has been input (S19). The input of the new fourth information is performed, for example, by an analyst who performs analysis using the association model using the input device 1. The new fourth information may be a numerical value that changes the value of the purchase intention element included in the association model.

[0104] When new fourth information is input (YES in S19), the prediction unit 214 calculates a predicted value of the first information when the new fourth information is input to the association model (S20: first prediction step). For example, the prediction unit 214 inputs the new fourth information and calculates predicted values of each feature element included in the first information under predetermined constraint conditions. The prediction unit 214 causes the display unit 22 to present a prediction result indicating the calculated predicted value (S21). Note that the prediction unit 214 may calculate a predicted value of the third information under predetermined constraint conditions based on the new fourth information, and further calculate a predicted value of the second information under predetermined constraint conditions using the new fourth information and the third information calculated based on the new fourth information.

[0105] As described above, in the evaluation support method according to Embodiment 2, when the characteristics of the evaluation target change, it is possible to predict how the purchase intention changes by using the created association model and the new first information. Further, in the evaluation support method according to Embodiment 2, when the purchase intention of the evaluation target is changed, it is possible to predict how the characteristics of the evaluation target change under predetermined constraint conditions by using the created association model and the new fourth information.

[0106] By checking the output prediction results, designers and the like of products can analyze the relationship between changes in what characteristics and the purchase intention of consumers, etc., and utilize it for product design. 〔Embodiment 3〕

[0107] FIG. 7 is a flowchart showing an example of the flow of the evaluation support method according to Embodiment 3. As shown in FIG. 7, the evaluation support method according to Embodiment 3 includes a second prediction step (S29). The second prediction step is a step of calculating a predicted value of a specified prediction target from the new information based on the new information indicating at least any one of the first information, the second information, the third information, and the fourth information newly input to the association model and the information specifying the prediction target. Further, the prediction target is at least any one of the information not included in the new information among the first information, the second information, the third information, and the fourth information.

[0108] FIG. 8 is a block diagram showing the configuration of the evaluation support system 100B according to Embodiment 3. As shown in FIG. 8, the evaluation support system 100B includes an input device 1B and a model creation device 2B. The input device 1B receives input of new information by an analyst who performs analysis using the association model. The model creation device 2B includes a control unit 21B, a display unit 22, and a storage unit 23. The control unit 21B is different from the control unit 21A according to Embodiment 2 in that it includes a prediction unit 214B instead of the prediction unit 214.

[0109] Hereinafter, the flow of the evaluation support method according to Embodiment 3 will be described with reference to FIGS. 7 and 8. Note that the processes of steps S21 to S26 in FIG. 7 are the same as the processes of steps S1 to S6 described with reference to FIG. 1 in Embodiment 1, and thus the description thereof will be omitted.

[0110] After the association model is created by the model creation unit 213 and displayed on the display unit 22, new information can be input to the input device 1B by an analyst who performs analysis using the association model. The new information is any one of new first information to fourth information. Information indicating which element among the elements included in the association model the new information corresponds to is associated with the new information.

[0111] The new first information may be a numerical value that changes the value of any one of the feature elements included in the association model. The new second information may be a numerical value that changes the value of any one of the impression elements included in the association model. The new third information may be a numerical value that changes the value of any one of the three elements of the attitude included in the association model. The new fourth information may be a numerical value that changes the value of the purchase intention element included in the association model. Note that the new information may be a plurality of new first information to fourth information.

[0112] After presenting the association model in S26, when new information is input to the input device 1B, the prediction unit 214B acquires the new information (YES in S27). When acquiring the new information, the prediction unit 214B receives a designation of a prediction target via the input device 1B. An analyst or the like can input information for designating a prediction target to the input device 1B. The prediction target is a target predicted based on the association model and the new information. At least one of the information other than the information input as the new information among the first information to fourth information, in other words, the information not included in the new information, can be designated as the prediction target.

[0113] When the prediction unit 214B receives information specifying a prediction target (YES in S28), it calculates a predicted value of the prediction target specified in S28 when new information is input into the association model (S29: second prediction step).

[0114] For example, a case where new fourth information is input as new information and the first information, the second information, and the third information are specified as prediction targets will be described. In this case, the prediction unit 214B inputs the new fourth information into the association model, and calculates predicted values of each feature element included in the first information, each impression element included in the second information, and the three elements of the attitude included in the third information under predetermined constraint conditions.

[0115] Note that any element included in the first to fourth information may be specified as a prediction target. For example, all elements included in the first information may be specified as prediction targets, or some elements of the first information may be specified as prediction targets. Specifically, the prediction unit 214B may calculate predicted values of all elements included in the first information based on the association model and the input new information. Further, the prediction unit 214B may calculate a predicted value of any element included in the feature elements of the first information based on the association model and the input new information.

[0116] The prediction unit 214B causes the display unit 22 to present a prediction result indicating the calculated predicted value (S30). As described above, in the evaluation support method according to Embodiment 3, when any of the first to fourth information of the evaluation target changes by using the created association model and new information, it is possible to predict how the information or element specified as the prediction target changes. A product designer or the like can analyze how much a change in a certain information or element included in the association model affects other information or elements, etc. by checking the output prediction result, and can use it for product design.

[0117] Note that in Embodiment 3, after the association model is presented in S26, the model creation device 2B or an external device communicably connected to the model creation device 2B may store the association model. In this case, when the model creation device 2B acquires new information via the input device 1B, the model creation device 2B may start the processing after S27.

[0118] 〔Example of Realization by Software〕 The functions of the evaluation support systems 100, 100A, and 100B (hereinafter referred to as "systems") are programs for causing a computer to function as the system, and can be realized by programs for causing a computer to function as each control block of the system (particularly each part included in the control units 21, 21A, and 21B).

[0119] In this case, as hardware for executing the above program, the system includes a computer (including a tablet terminal and a smartphone) having at least one control device (for example, a processor) and at least one storage device (for example, a memory). By executing the above program with this control device and storage device, each function described in each of the above embodiments is realized.

[0120] The above program may be recorded on one or more computer-readable recording media, not temporarily. This recording media may or may not be provided in the above device. In the latter case, the above program may be supplied to the above device via any wired or wireless transmission medium.

[0121] Also, part or all of the functions of each of the above control blocks can also be realized by a logic circuit. For example, an integrated circuit in which a logic circuit functioning as each of the above control blocks is formed is also included in the scope of the present invention. In addition to this, for example, it is also possible to realize the functions of each of the above control blocks by a quantum computer.

[0122] In addition, each process described in each of the above embodiments may be executed by AI (Artificial Intelligence). In this case, the AI may operate in the above control device, or may operate in another device (for example, an edge computer or a cloud server, etc.).

[0123] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope shown in the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention.

[0124] 〔Summary〕 The evaluation support method according to Aspect 1 of the present invention includes a presentation step of presenting an evaluation target having a plurality of characteristics to a subject, and a question including a question sentence related to each of the plurality of characteristics and answer options for the question sentence with respect to the evaluation target, wherein at least one evaluation word related to the evaluation target is included in at least one of the question sentence and the answer options. An acquisition step of a computer acquiring an answer of the subject to a question group including a plurality of such questions, and a computer, based on each of the plurality of characteristics, a word indicating a factor or component common to the evaluation words included in the question group, and the answer of the subject, (1) first information indicating each of the plurality of characteristics, (2) second information regarding an impression received by the subject from the evaluation target based on each of the plurality of characteristics, (3) third information regarding the attitude of the subject according to the impression, and (4) fourth information regarding the purchase intention of the subject according to the impression or the attitude. A model creation step of creating a correlation model including, wherein the evaluation words include words related to the impression, words related to the attitude, and words related to the purchase intention.

[0125] The evaluation support method according to Aspect 2 of the present invention may further include, in the above Aspect 1, an extraction step in which a computer extracts a word indicating a factor or component common to the evaluation words included in the question group for each layer indicating the nature of the word for classifying the evaluation words.

[0126] In the evaluation support method according to Aspect 3 of the present invention, in the above Aspect 1 or 2, the words related to the attitude may include words indicating the subject's recognition of the evaluation target according to the impression, the feelings that the subject has towards the evaluation target, and the behavioral tendency of the subject towards the evaluation target.

[0127] In the evaluation support method according to Aspect 4 of the present invention, in any one of the above Aspects 1 to 3, based on the newly input first information or the newly input fourth information to the association model, a first prediction step of calculating a predicted value of the information that was not newly input among the first information and the fourth information from the newly input first information or the newly input fourth information may be included.

[0128] In the evaluation support method according to Aspect 5 of the present invention, in any one of the above Aspects 1 to 4, based on new information indicating at least any one of the newly input first information, second information, third information, and fourth information to the association model, and information specifying a prediction target, a second prediction step of calculating a predicted value of the specified prediction target from the new information is included, and the prediction target may be at least any one of the first information, second information, third information, and fourth information that is not included in the new information.

[0129] In the evaluation support method according to Aspect 6 of the present invention, in any one of the above Aspects 1 to 5, the plurality of features may be either a measurement result related to the nature of the evaluation target or an evaluation result obtained by evaluating the features based on a predetermined standard.

[0130] The evaluation support system according to Aspect 7 of the present invention includes questions presented to a subject about an evaluation target having a plurality of features, the questions including question sentences related to each of the plurality of features and answer options for the question sentences, and at least one evaluation word related to the evaluation target is included in at least one of the question sentences and the answer options. An acquisition unit that acquires the subject's answers to a question group including a plurality of such questions, and based on each of the plurality of features, a word indicating a factor or component common to the evaluation words included in the question group, and the subject's answers, (1) first information indicating each of the plurality of features, (2) second information regarding the impression received by the subject from the evaluation target based on each of the plurality of features, (3) third information regarding the subject's attitude according to the impression, and (4) fourth information regarding the subject's purchase intention according to the impression or the attitude. A model creation unit that creates a correlation model including, the evaluation words include words related to the impression, words related to the attitude, and words related to the purchase intention.

Explanation of Signs

[0131] S1, S11, S21 Presentation step S2, S12, S22 Acquisition step S3, S13, S23 Extraction step S5, S15, S25 Model creation step S18 First prediction step S20 First prediction step S29 Second prediction step

Claims

1. A presentation step of presenting an evaluation target having a plurality of features to a subject; an acquisition step in which a computer acquires answers from the subject to a group of questions including a question sentence related to each of the plurality of features of the evaluation target and answer options to the question sentence, wherein at least one of the question sentence and the answer options includes at least one evaluation term related to the evaluation target; and a model creation step in which a computer creates an association model based on each of the plurality of characteristics, words indicating factors or components common to the evaluation words included in the group of questions, and the answers of the subject, the association model including: (1) first information indicating each of the plurality of characteristics; (2) second information on the impression the subject has of the evaluation object based on each of the plurality of characteristics; (3) third information on the attitude of the subject according to the impression; and (4) fourth information on the purchase intention of the subject according to the impression or the attitude, The evaluation words include words related to the impression, words related to the attitude, and words related to the purchase intention. Evaluation support methods.

2. 2. The evaluation support method according to claim 1, further comprising an extraction step in which a computer extracts words indicating factors or components common to the evaluation words included in the group of questions for each layer indicating the properties of words for classifying the evaluation words.

3. The words related to the attitude are: The evaluation support method according to claim 1 , further comprising words indicating the subject's perception of the evaluation object according to the impression, the subject's feelings toward the evaluation object, and the subject's behavioral tendency toward the evaluation object.

4. 2. The evaluation support method according to claim 1, further comprising a first prediction step of calculating a predicted value of information that has not been newly input of the first information and the fourth information from the newly input first information or the newly input fourth information based on the newly input first information or the newly input fourth information that has been newly input to the association model.

5. a second prediction step of calculating a prediction value of the specified prediction target from new information based on new information indicating at least one of the first information, the second information, the third information, and the fourth information newly input to the association model and information specifying a prediction target; The prediction target is at least any one of the first information, the second information, the third information, and the fourth information that is not included in the new information. The evaluation support method according to claim 1 .

6. The evaluation support method according to claim 1 , wherein the plurality of features are either measurement results relating to properties of the evaluation object or evaluation results obtained by evaluating the features based on a predetermined criterion.

7. an acquisition unit that acquires answers from the subject to a group of questions including a question sentence related to each of a plurality of characteristics of an evaluation target that is presented to the subject and an answer option to the question sentence, the question sentence and at least one of the answer options include at least one evaluation term related to the evaluation target; and a model creation unit that creates an association model based on each of the plurality of features, words indicating factors or components common to the evaluation words included in the group of questions, and the answers of the subject, the model including: (1) first information indicating each of the plurality of features; (2) second information on the impression the subject has of the evaluation object based on each of the plurality of features; (3) third information on the attitude of the subject according to the impression; and (4) fourth information on the purchase intention of the subject according to the impression or the attitude, The evaluation words include words related to the impression, words related to the attitude, and words related to the purchase intention. Evaluation support system.

Images