JP4953079B2 - Information processing apparatus, Kansei modeling system, Kansei modeling method and program - Google Patents

Description

  The present invention relates to an information processing apparatus, a sensibility modeling system, a sensibility modeling method, and a program for quantifying a user's preference for a product and associating the user's preference with a sensibility target.

  At present, a wide variety of products, information, and contents have been distributed to society due to the spread of the Internet and the globalization of society. This makes it possible for people to choose a product that suits their interests from among many options.

  For example, Japanese Patent Application Laid-Open No. 2004-164374 (Patent Document 1) discloses a data processing device that uses a rough set and uses a request level composed of message elements to obtain message element conditions. Japanese Patent Laid-Open No. 2004-213463 (Patent Document 2) uses a generalized distribution table and a rough set to create a probability criterion for evaluating the strength of a rule. A data processing method for extracting a rule set is disclosed. Information processing using rough sets is described in, for example, Norihiko Mori, Hideo Tanaka, Katsuo Inoue, “Rough Sets and Kansei”, Kaibundo Publishing, 2004 (Non-Patent Document 1).

  Although the above-described technology has been proposed, it is not easy to show the user the relevance between numerical data and palatability simply by presenting or evaluating the palatability of the user. There was a problem.

In other words, the presence of a large group of sensibility objects that are beneficial to the user sometimes immerses the sensibility object such as a product suitable for each user's interest in the sea of other sensibility objects and cannot find the product that he or she desires There was a problem of reducing opportunities for me to know new products.
JP 2004-164374 A JP 2004-213463 A Norihiko Mori, Hideo Tanaka, Katsuo Inoue, “Rough Assembly and Sensitivity”, Kaibundo Publishing, 2004

  The present invention has been made in view of the above-mentioned problems of the prior art and solves the problem of overlooking sensitivity objects such as products, information, and contents that match the sensitivity of each user, such as interests. To provide information processing apparatus, sensitivity modeling system, sensitivity modeling method and program for modeling relationship between interests of goods and attributes of products, and using the model to support selection of sensitivity targets of each user With the goal.

That is, according to the present invention,
An information processing apparatus that models a sensitivity to a sensitivity object,
Interest information acquisition means for acquiring and registering interest information about the sensitivity object;
Calculating a rough set of the interest information using the attribute value of the interest information acquired by the interest information acquiring means, and registering it in the rough set database in association with the user identification value;
Interest information modeling for calculating a preference measure and an attribute distribution measure indicating user preference using the interest information registered in the rough set database and the sensitivity target attribute and the sensitivity target attribute value stored in the attribute database Means,
There is provided an information processing apparatus including: a sensitivity map creating means for mapping the sensitivity object to the preference scale created by the interest information modeling means and the attribute distribution scale to create a sensitivity map.

  In the present invention, the Kansei map creating means includes categorizing means for categorizing the Kansei objects from the distribution of the interest information on the Kansei objects using a threshold set for each scale. Can do. In the present invention, the interest information acquisition means can acquire the interest information via a network and pass the acquired interest information to the rough set attribute analysis means. Furthermore, the categorizing means can categorize the sensitivity objects registered in the attribute database using the interest information registered in the rough set database.

Moreover, according to the present invention,
A sensitivity modeling method in which a computer models a sensitivity to a sensitivity object, the sensitivity modeling method comprising:
Acquiring and registering interest information about the sensitivity object;
Calculating a rough set of the interest information using the acquired attribute value of the interest information, registering it in the rough set database in association with a user identification value;
Calculating a preference scale and an attribute distribution scale indicating user preference using the interest information registered in the rough set database and the sensitivity target attribute and the sensitivity target attribute value stored in the attribute database;
There is provided a sensitivity modeling method for executing the step of mapping the sensitivity object to the preference measure and the attribute distribution measure to create a sensitivity map.

  The step of creating the sensitivity map of the present invention includes categorizing the sensitivity object using a threshold value set for each scale from the distribution of the interest information about the sensitivity object. it can. The step of acquiring and registering the interest information may include the step of acquiring the interest information via a network and passing it to determine an attribute of the rough set. The categorizing step categorizes the sensitivity object registered in the attribute database using the interest information registered in the rough set database, and the sensitivity modeling method includes the sensitivity object stored in the attribute database. Information can be provided.

  Furthermore, according to the present invention, a computer-executable program for causing a computer to function as each of the above-described means is provided.

Moreover, according to the present invention,
Network,
The information processing apparatus according to any one of claims 1 to 4,
A sensibility modeling system including a network client connected to the network and transmitting interest information about a sensibility object to the information processing apparatus is provided.

  Hereinafter, although this invention is demonstrated based on specific embodiment, this invention is not limited to embodiment mentioned later.

<Section 1: Modeling the relationship between the user's interest (Kansei) and Kansei objects>
In this embodiment, in order to specifically explain the relationship between the user's sensibility and the attribute of the sensibility target, the user's sensibility is the interest in the product, the user's sensibility target is the product, and the relationship between the product attributes I do. Then, using a relationship between the user's sensitivity (interest) and the attribute value of the sensitivity object, the estimation of the product suitable for the user's interest and the visualization thereof will be specifically described. In the present embodiment, Kansei modeling is performed by (1) obtaining the relationship between user interests and non-interests and attributes of each product, (2) extracting a set of product attributes that the user is interested in, and (3) users. This is done using a three-step process of modeling the relationship between the interest of the product and the attribute of the product.

(1-1) Acquisition of relationship between user's interests and non-interests and attributes of each product In order to model user's interests and product attributes and relationships, the model is based on "how each user is Data for each user needs to be acquired. FIG. 1 is a diagram showing an embodiment of an attribute correspondence table 100 showing the relationship between the interests and non-interests of the user x and the attributes of _five products (I ₁ , I ₂ , I ₃ , I ₄ , I ₅ ). is there. The data shown in FIG. 1 is accumulated by a server included in the sensibility modeling system by using a questionnaire to the user or using questions about product characteristics displayed on a web page. This data is used to create a rough set as described below.

(1-2) Extraction of Set of Product Attributes with Interest of User In the present embodiment, the attribute of the product with interest of user x is calculated using calculation of a rough set theory decision rule (Non-Patent Document 1). Set: Cx (hereinafter referred to as a set of interest attributes) is extracted. The following is the process of deriving the set of interest attributes: Cx. The difference between the attribute / attribute value of the product that user x is interested in and the product that is not interested in is derived. For example, in the table of FIG. 1, each product (I ₁ , I ₂ , I ₄ , I ₅ ) that the user x replied with interest and a product (I ₃ ) that replied with no interest are the attributes shown in FIG. -Differentiated by the data structure in which the attribute value is registered, and using the rough set theory, it becomes possible to find the attribute / attribute value in the product that user x is interested in.

In the extraction of the attribute set of the product, the attribute value identifying “interested / none” shown in FIG. 2 is used to perform an operation on the set variable for the rough set shown in FIG. Derive Cx. For example, the interest attribute: Cx possessed by the product (I ₁ , I ₂ , I ₄ , I ₅ ) that the user x has shown interest can be derived by the calculation of the following formula (1).

<Section 2: Modeling the relationship between user sensibility and sensibility target attributes>
In this embodiment, in order to derive the relationship between the user's interest (Kansei) and the attributes of the product that is the sensibility target, a known product that is considered to reflect each user's interest (e.g., possessed, interest) It is necessary to know what the configuration of (shown) is. In the present embodiment, the term “known product” includes not only the possessed item but also the item that shows interest even if it is not actually possessed (for example, a product that is included in Wishlist). Define. In the following, reference will be made to the same meanings unless otherwise indicated.

  For example, if the product (clothes) possessed by user A contains many interest attributes: (color (blue), shape (shirt)), for user A, the interest attributes: (color (blue), shape (shirt)) ) Can be inferred to be frequently selected products (hereinafter referred to as standard products). Further, if the product (clothes) possessed by the user A includes many attributes / attribute values: color (blue), the user A can be considered to have a high degree of interest in the attributes / attribute values: color (blue). . Therefore, in the embodiment, the following two amounts of popularity and personality are calculated, and the relationship between the user's interest and the product attribute is calculated by calculating the relationship between this information and the attribute / attribute value of the user's known product. Model relationships.

  In the present embodiment, the degree of popularity is defined as the appearance frequency of each interest attribute in the product owned by the user. The personality degree is defined as a value of an attribute value for each attribute in the product owned by the user. Hereinafter, the calculation processing of each information amount will be specifically described.

  In the present embodiment, the degree of popularity can be defined by the following formula (2).

More specifically, the degree of popularity Po (Cxy) of Cx1 (b ₁ , a ₁ ) and Cx2 (b ₁ , c ₁ ) of the set of interest attributes of user x calculated from FIG. 1 and FIG. Calculated as follows: For example, since Cx1 (b ₁ , a ₁ ) is included only in the product I ₁ , nCxy is 1, and since the number of products possessed by the user x is 5, the degree of popularity is 0.2. .

  In the present embodiment, the degree of personality can be defined by the following formula (3).

For example, the personality degree Pa (B _b1 ) of the attribute value b ₁ of the attribute B can be calculated as follows. However, in this example, the number of attributes Kv in all the products possessed by the user x is calculated as five types, and the attribute B types are calculated as three types (b ₁ , b ₂ , b ₃ ).

<Section 3: Modeling the relationship between user interests and product attributes>
In the present embodiment, the relationship between the user's interest in the known product of the user and the attribute of the product is modeled by quantifying the correspondence between the attribute / attribute value of each user's known product, the degree of popularity and personality of each user. Turn into. The degree of popularity is used to calculate the degree of popularity reflection, which is an attribute distribution measure for product attributes. In the present embodiment, the degree of personality is used to calculate a degree of personality reflection, which is a user preference measure indicating the strength of user preference.

In this study, the relationship between the user's popularity degree and the attribute / attribute value of the product is calculated as the popularity reflection degree: CPo (Co _j ) by the method shown in the following formula (4).

The relationship between the user's personality level and the product attribute / attribute value is calculated as a personality reflection level CPa (Co _j ) by the method shown in the following equation (5).

  In this embodiment, the interest of each user and the attribute of the product are obtained by associating each product with a two-dimensional space (hereinafter referred to as “Kansei map”) in which the popularity reflection is on the vertical axis and the personality reflection is on the horizontal axis. Visualize and express the relationship. FIG. 3 shows an embodiment of the sensitivity map 300. In the embodiment shown in FIG. 3, in order to describe this embodiment more specifically, it is assumed that the sensitivity target is a product, more specifically, clothing that strongly reflects the user's preference. The sensitivity map 300 shown in FIG. 3 analyzes the relationship between each user's interest and existing products by dividing them into the following four categories. The four categories are defined as follows:

(1) Standard Product A product having a popularity reflection degree and a personality reflection degree higher than the set threshold value is considered to be a standard commodity 306 that is frequently selected for the user.
(2) Rare Products A rare product 302 having a popularity reflection level lower than a set threshold value and a personality reflection rate higher than a set threshold value is considered to be very important for this user even if it is a small number.
(3) Base product The base product 308 having a popularity reflection degree higher than the set threshold and a personality reflection degree lower than the set threshold is considered to be a necessity for the user, although he is not interested in occupation or life.
(4) Single-item products Single-item products 304 whose both popularity reflection degree and personality reflection degree are lower than the set threshold values are not measures of popularity reflection degree and personality reflection degree. It can be classified as a terminal product as an influence by a program, a present, or the like.

  Each threshold can be determined experimentally by data collection experiments using appropriate investigators or volunteers in order to classify the user's sensibilities. For example, the types of goods, such as clothing, passenger cars, watches, etc. It can be decided according to such as.

<Section 4: Kansei modeling system>
FIG. 4 shows an embodiment of the Kansei modeling system 400 of this embodiment. The sensibility modeling system 400 shown in FIG. 4 can be implemented including an information processing apparatus 402 implemented as a personal computer, a workstation, or a dedicated server. The information processing apparatus 402 can be implemented including an appropriate operating system (hereinafter referred to as OS) such as WINDOWS (registered trademark) XP, WINDOWS (registered trademark) 200X server, MACOS X, and the like. Below, the program described in programming languages, such as Java (trademark), C ++, Perl, RUBY, is performed. Further, when the information processing apparatus 402 is configured as a web server, a server program such as a CGI program, Servlet, or Appache is installed, the interest information is acquired via the network 424, and the sensitivity map is transmitted to the user via the network 424. Can be sent to.

  In the specific example of the present embodiment, the information processing apparatus 402 is mounted on Mac OS X using Java (registered trademark) 5.0 + PostgreSQL 8.2 + Jakata Tomcat 5.5. As shown in FIG. 4, the sensibility modeling system 400 includes an interest information acquisition unit 404, a rough set attribute analysis unit 406, and an interest information modeling unit 408. The interest information acquisition unit 404 receives interest information from the network 424 or the mouse 422 and the keyboard 420 and registers each interest information in a buffer memory or the like in association with the user identification value. The rough set attribute analysis unit 406 acquires the interest information acquired by the interest information acquisition unit 404, extracts the attribute value of the rough set, creates a rough set, and registers it in the rough set DB 414 in association with the user identification value. To do. For this reason, the rough set attribute analysis unit 406 has a relational database function.

  In addition, the sensibility modeling system 400 manages an attribute database (attribute DB) 416 that registers a product attribute of the product identification value with respect to the product identification value. In the described embodiment, the attribute database 416 is implemented as a product database, and registers attribute values of commercial objects such as clothes, cars, foods, health foods, and contents. In this embodiment, it is possible to register attribute values that characterize the general public's sensuous and psychological behaviors or actions as well as products.

  The interest information modeling unit 408 uses the interest information accumulated in the rough set database (rough set DB) 414 and the product attributes stored in the attribute DB 416, and uses the above formulas (2) to (7). A value that gives a measure of preference for a given product is calculated, and the calculated personality reflection degree and popularity reflection degree are registered in the rough set DB 414 in association with the user identification value and the product identification value, and a Kansei model data structure is generated. To do.

  The sensitivity map creation unit 410 corresponds to the user's interest information, acquires the personality reflection degree and popularity reflection degree calculated for each product, normalizes each coordinate value with respect to the maximum value and the minimum value, The center of the product image is made to coincide with the coordinate position corresponding to the values of the popularity reflection degree and the personality reflection degree, which are the preference scales calculated for the identification values, and a map displayed with an appropriate reduced scale is created. This map can be in various formats, for example, GIF, BMP, PNG, JPEG, JPEG2000, PDF, etc.

  The input / output interface unit 412 displays the created sensitivity map on the display device 418, and displays a product that matches the user's preference for the user. The sensitivity modeling system 400 can also display a sensitivity map created for a network client (not shown) used by the user via the network 424. In the case of this embodiment, the information processing apparatus 402 has a CGI program for creating a sensitivity map. In addition, the network client can mount a web browser such as Netscape Navigator (trademark), internet Explorer (trademark), Mozilla, and Opera, and can display the sensitivity map using the HTTP protocol. Note that the term “sensitivity map” used in the present embodiment reflects preferences such as cars, foods, clothing, hobbies, music, books, movies, contents, etc. It means a data structure in which attribute values are mapped, and there is no particular limitation on the object.

  FIG. 5 is a flowchart of an embodiment of processing executed by the sensibility modeling system 400 of the present embodiment. The process of the sensibility modeling system 400 starts from step S500, presents a product image to the user in step S501, acquires the presence or absence of interest for each attribute, and registers it as interest information in a buffer memory or the like. In step S502, a rough set is created from interest information indicating the interest, and registered in the database in association with the product image identification value and the user identification value.

  In step S503, the degree of popularity and the degree of personality are calculated from the information registered in the database, and the degree of popularity reflection and the degree of personality reflection are calculated and registered in the database in association with the user identification value. In step S504, the center of the product image is made to coincide with the coordinate position of the personality reflection degree and the popularity reflection degree of the product image using the popularity reflection degree and the personality reflection degree as coordinate axes, and a map is created. In this case, it is also possible to create a line or box corresponding to each threshold value and notify the user of which category range the preference is included in.

  In step S506, the map is displayed on the display device, or the map is transmitted via the network, and the map is presented to the user. In step S507, the series of processing ends.

  FIG. 6 illustrates an embodiment of a graphical user interface (GUI) displayed on the display device 418 to obtain interest information in this embodiment. The GUI 600 shown in FIG. 6 displays a field 602 for displaying a product image and a field 604 for displaying a product attribute. The user confirms the displayed product attributes, and if interested, selects the radio button 606 displayed as “Interested” and clicks the “Enter” button 610 to enter the interest modeling information into the Kansei modeling system. Register. Note that the product image repeats the same process using a plurality of different product images as samples in order to obtain as much preference as possible for a specific user.

<Section 5: Kansei modeling experiment with volunteers as test users>
Using the sensibility modeling system 400 described above, seven volunteers are subject users, and whether or not they are interested in 94 types of clothes randomly presented from 949 types of images is input to obtain interest information. The results were obtained through analysis using a Kansei modeling system. In this experiment, 94 types of clothes were processed as products held by the subject user (hereinafter referred to as existing products).

  FIG. 7 is an embodiment of an interest attribute set indicating the preference of the subject user A calculated in the present embodiment. The interest attribute set shown in FIG. 7 is collected by the Kansei modeling system 400 using the GUI 600 shown in FIG. In the embodiment shown in FIG. 7, the element values of the interest attribute set of the subject user A are extracted (13 sets out of 41 sets). In the case of the subject user A, as shown in FIG. 7, the color interest attribute has nine attributes, the interest attribute regarding the arm shape (sleave) has three attributes, and the subject user A has the color and arm shape. It is shown that it has strong palatability.

  FIG. 8 shows excerpts of the degree of popularity of subject user A (5 of 41 sets) (FIG. 8A) and excerpts of personality (FIG. 8B). For example, the degree of popularity of the interest set Ca7 (color = Black, neck = Hood) of the subject user A shown in FIG. 7 is 0.021052631578947368 from FIG. 8A, and the degree of popularity is the degree of popularity of the interest attribute Ca21. It is shown to be higher. FIG. 8B shows the personality degree of the subject user A calculated from the interest information. This test subject user A has a personality degree (preference) of attributes such as no waist accents (None-Waist) and a cut-and-sew (Cut_and_Swen) attribute, and a chest mark (Mark) It is shown that the degree of personality (preference) for attributes such as short sleeves (Half) is low.

  In FIG. 9 and FIG. 10, the popularity reflection degree and the personality reflection degree for each existing product are calculated for the interest models of the subject user A and the subject user B, respectively, and presented as a sensitivity map.

  As shown in FIG. 9, the subject user A has a hooded cut-and-sew outfit on the base product. On the other hand, in the test subject user B shown in FIG. 10, many blue clothes are located as base products. The product classifications shown in FIGS. 9 to 12 were set by analyzing the personality reflection degree and popularity reflection degree for 94 types of images and determining each threshold value so as to be appropriately distributed. As shown in FIG. 9 and FIG. 10, it is mapped that the weight (personality) of the attribute that the user regards is different for each user. In addition, the difference is clearly reflected in the selection frequency (popularity) of the products that the user is interested in, and the Kansei modeling system 400 succeeds in appropriately modeling and classifying the Kansei of the user. Is shown.

  FIG. 11 and FIG. 12 calculate the popularity reflection degree and the personality reflection degree for unknown products unknown to the test subject user A and the test subject user B, respectively, and display the product at the corresponding coordinate positions and display them as a sensitivity map. The results are shown. Each unknown product in FIG. 11 and FIG. 12 is arranged at a position corresponding to the known product shown in FIG. 9 and FIG. 10, that is, a position of a standard product, a base product, a single product, and a rare product. Recognize. More specifically, in FIG. 11, for subject user A, hooded cut-and-sew clothes are gathered at the position of the base product, and for subject user B, blue clothing is gathered as the base product. It is shown.

  As shown in FIGS. 9 to 12, in this embodiment, the degree of popularity and personality of each user are calculated, and the degree of popularity reflection and personality reflection of an unknown product that reflects the user's interest is calculated. Is possible. And each user can search for the unknown goods suitable for the said user's interest and preference about the goods in which the preference was classified on the basis of the calculated popularity reflection degree and personality reflection degree.

  As described above, according to the present invention, the relationship between the user's interest and the product attribute is modeled by calculating the degree of popularity and the degree of personality, and selecting an unknown product for each user using the model. Can be supported. That is, according to the present invention, it is possible to provide an information processing apparatus, a sensitivity modeling system, a sensitivity modeling method, and a program capable of evaluating an unknown product suitable for the user's interest while reflecting the user's interest.

  In addition, the sensitivity modeling system of the present invention not only presents unknown products to the user, but also provides a product development direction that reflects the preferences of a plurality of users.

  The above-described functions of the present invention are described in object-oriented programming languages such as C ++, Java (registered trademark), Java (registered trademark) Beans, Java (registered trademark) Applet, Java (registered trademark) Script, Perl, and Ruby. It can be realized by a device executable program, and can be stored and distributed in a device-readable recording medium.

  The present invention has been described with the embodiment. However, the present invention is not limited to the embodiment, and other embodiments, additions, modifications, deletions, and the like can be conceived by those skilled in the art. It can be changed, and any aspect is within the scope of the present invention as long as the effects and effects of the present invention are exhibited.

Interest of the user x, illustrates an embodiment of a table showing the attributes of the relationship between the non-interest and five items _{(I 1, I 2, I} 3, I 4, I 5). FIG user with each item answered curious _{(I 1, I 2, I} 4, I 5) that there is no interest and answer products (I _3), shows the data structure for distinguishing. The figure which showed embodiment of the sensitivity map. The figure which showed embodiment of the sensitivity modeling system of this embodiment. The flowchart of embodiment of the process which the Kansei modeling system of this embodiment performs. The figure which showed embodiment of the graphical user interface (GUI) displayed on a display apparatus in order to acquire interest information in this embodiment. The figure which showed embodiment of the interest attribute set which shows the test subject user's A preference computed by this embodiment. The figure which showed embodiment of the test subject user's A popularity degree extract (FIG. 8 (a)) and the excerpt of a personality degree (FIG.8 (b)). The figure which showed embodiment of the sensitivity map which calculated the popularity reflection degree and personality reflection degree about each existing goods with respect to the test subject user's A interest model. The figure which showed embodiment of the sensitivity map which calculated the popularity reflection degree and personality reflection degree about each existing product with respect to the test subject user's B interest model. The figure which showed other embodiment of the sensitivity map which calculated the popularity reflection degree and personality reflection degree about each existing goods with respect to the test subject user's A and B interest model. The figure which showed other embodiment of the sensitivity map which calculated the popularity reflection degree and personality reflection degree about each existing goods with respect to the test subject user's B interest model.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Attribute correspondence table, 200 ... Data structure, 300 ... Kansei map, 400 ... Kansei modeling system, 402 ... Information processing apparatus, 404 ... Interest information acquisition part, 406 ... Rough set attribute analysis part, 408 ... Interest information modeling 410, Kansei map creation unit, 412 ... I / O interface unit, 414 ... Rough set DB, 416 ... Attribute DB, 418 ... Display device, 420 ... Keyboard, 422 ... Mouse, 424 ... Network, 600 ... GUI

Claims (10)

An information processing apparatus that models a sensitivity to a sensitivity object,
Interest information acquisition means for acquiring and registering interest information about the sensitivity object;
Calculating a rough set of the interest information using the attribute value of the interest information acquired by the interest information acquiring means, and registering it in the rough set database in association with the user identification value;
Interest information modeling for calculating a preference measure and an attribute distribution measure indicating user preference using the interest information registered in the rough set database and the sensitivity target attribute and the sensitivity target attribute value stored in the attribute database Means,
An information processing apparatus comprising: a sensitivity map creating means for mapping the sensitivity object to the preference scale and the attribute distribution scale created by the interest information modeling means to create a sensitivity map.   The sensibility map creation means includes categorization means for categorizing the sensibility objects using a threshold value set for each scale from the distribution of the interest information about the sensibility objects. The information processing apparatus described.   The information processing apparatus according to claim 1, wherein the interest information acquisition unit acquires the interest information via a network and passes the acquired interest information to the rough set attribute analysis unit.   The information processing apparatus according to claim 2 or 3, wherein the categorizing unit categorizes the sensitivity object registered in the attribute database using the interest information registered in the rough set database. A sensitivity modeling method for modeling a sensitivity to a sensitivity object executed by a computer, the sensitivity modeling method comprising:
Acquiring and registering interest information about the sensitivity object;
Calculating a rough set of the interest information using the acquired attribute value of the interest information, registering it in the rough set database in association with a user identification value;
Calculating a preference scale and an attribute distribution scale indicating user preference using the interest information registered in the rough set database and the sensitivity target attribute and the sensitivity target attribute value stored in the attribute database;
A method for creating a sensitivity map by mapping the sensitivity object to the preference scale and the attribute distribution scale and creating a sensitivity map.   6. The step of creating the sensitivity map includes categorizing the sensitivity object from a distribution of the interest information about the sensitivity object using a threshold value set for each of the scales. The sensibility modeling method described.   The Kansei modeling method according to claim 5 or 6, wherein the step of acquiring and registering the interest information includes the step of acquiring the interest information via a network and passing the interest information to determine an attribute of the rough set. .   The categorizing step categorizes the sensitivity object registered in the attribute database using the interest information registered in the rough set database, and the sensitivity modeling method includes the sensitivity stored in the attribute database. The sensitivity modeling method according to claim 6 or 7, wherein target information is provided.   The computer-executable program for functioning a computer as each means of any one of Claims 1-4. Network,
The information processing apparatus according to any one of claims 1 to 4,
A sensibility modeling system including a network client connected to the network and transmitting interest information about a sensibility target to the information processing apparatus.

Images