Material recommendation device, material recommendation method, and program

Abstract

A material recommendation device is provided with: an input unit that receives input of information indicating a product type and an emotion language indicating a language that expresses an emotion; a material parameter extraction unit that extracts, on the basis of material parameter information in which a material used in a product belonging to the product category and a value of a sensory parameter indicating the degree of feeling felt by a person with respect to the material are associated with each other, the value of the sensory parameter for each material corresponding to the product category; an emotional language parameter extraction unit that extracts the value of the sensory parameter corresponding to the emotional language on the basis of emotional language parameter information in which the emotional language and the value of the sensory parameter are associated; and a recommended material determination unit that determines a material to be recommended to an emotional language on the basis of the value of the sensory parameter for each material extracted by the material parameter extraction unit and the value of the sensory parameter extracted by the emotional language parameter extraction unit.

Description

Technical Field

[0001] The present invention relates to a material recommendation device, a material recommendation method and a program.

[0002] This application claims priority based on Japanese Patent Application No. 2022-152385 filed in Japan on September 26, 2022, and the contents are incorporated herein by reference. Background Art

[0003] As an index used as a criterion for evaluation when selecting a product, for example, the characteristics and performance of the materials used in the product can be cited. In addition, how the material is felt by the five senses (sight, touch, hearing, taste, smell) and how the material appeals to the senses are also important indicators used as a criterion for evaluation when selecting a product.

[0004] In the past, it was difficult to accurately suggest materials that would give users the desired feeling. In order to recommend materials that correspond to the user's desired feeling, for example, the person in charge who wants to recommend materials needs to fully listen to the user. However, in this case, whether the material that gives users the desired feeling is appropriately recommended is also greatly influenced by the senses of the person in charge. For example, this is because even if the user seeks "comfortable" materials, the comfort level perceived by the user and the comfort level perceived by the person in charge may not necessarily be consistent.

[0005] To address such a problem, for example, a similar tactile material recommendation system described in Patent Document 1 stores parameter values ​​in advance, which indicate the correspondence between terms expressing tactile sensation (onomatopoeia) and physical properties of the material surface, and automatically suggests recommended materials for the input onomatopoeia based on the parameter values.

[0006] Prior art literature

[0007] Patent Literature

[0008] Patent Document 1: JP Patent No. 4469908 Summary of the invention

[0009] Problems to be solved by the invention

[0010] The similar tactile material recommendation system described in Patent Document 1 is a system that only responds to input based on onomatopoeias that represent tactile sensations. However, in reality, the impression that a user feels from a material is not necessarily the impression that is felt through touch. The impression that a user feels from a material sometimes includes impressions that are felt through multiple senses among the five senses, or impressions that are felt through all senses.

[0011] In addition, when a user expresses the impressions felt from a material in words, it is not limited to onomatopoeia that simply virtually expresses the impressions obtained through the five senses (e.g., "哗啦哗啦" (flapping sound), "呲溜呲溜" (slipping sound), etc.). It is not uncommon for a user to express the impressions felt from a material in words that express emotions. The so-called words that express emotions here are adjectives and the like used when numerically expressing human sensations in general sensory evaluations. Such adjectives also include, for example, "喜欢的" (likable), "古怪的" (peculiar), "舒爽的" (refreshing), and "有高级感的" (having a sense of luxury), which are adjectives that are difficult to capture in only specific sensations among the five senses. Due to the above situation, in the existing technology, there is a problem that it is difficult to recommend materials according to human emotions.

[0012] The present invention is proposed in view of such a situation, and provides a material recommendation device, a material recommendation method, and a program that can recommend materials according to human emotions.

[0013] Means for Solving the Problem

[0014] The content of the present disclosure includes the following embodiments.

[0015] [1] A material recommendation device, comprising: an input unit that receives an input of information representing a product category and an emotion word, where the emotion word represents a word that expresses an emotion; a material parameter extraction unit that extracts, based on material parameter information that correlates the materials used in products belonging to the product category with the values of sensory parameters representing the degree of sensation felt by a person for the materials, the values of the sensory parameters for each material corresponding to the product category input to the input unit; an emotion word parameter extraction unit that extracts, based on emotion word parameter information that correlates the emotion word with the values of the sensory parameters, the values of the sensory parameters corresponding to the emotion word input to the input unit; and a recommended material determination unit that determines, based on the values of the sensory parameters for each material extracted by the material parameter extraction unit and the values of the sensory parameters extracted by the emotion word parameter extraction unit, the material recommended for the emotion word input to the input unit.

[0016] [2] Based on the material recommendation device described in [1], the emotion word is a statement that is classified into an evaluative factor or an activity factor among the factors of the EPA (Evaluation, Potency, Activity) structure defined in the impression evaluation based on the SD (Semantic Differential) method.

[0017] [3] The material recommendation device according to [1] or [2] further comprises: a correction unit that corrects the value of the sensory parameter using a coefficient that takes into account a cross-modal effect of correlations between impressions perceived using a plurality of senses.

[0018] [4] Based on the material recommendation device described in [3], different values ​​are set in the coefficient according to the order in which impressions are felt through the multiple senses, and the correction unit obtains information representing the order and uses the coefficient corresponding to the obtained order to perform correction.

[0019] [5] Based on the material recommendation device described in [4], in the coefficient used to correct the value of the sensory parameter of each of the materials, a value of 1 is set in the case of the feeling felt first, a value greater than 0 and less than 1 is set in the case of the feeling felt later, and 0 is set in the case of other feelings.

[0020] [6] Based on the material recommendation device described in [4] or [5], in the coefficient used to correct the value of the sensory parameter of the emotional language, a value of 1 is set in the case of the feeling that produces the cross-modal effect, and a value of 0 is set in the case of the feeling that does not produce the cross-modal effect.

[0021] [7] Based on any one of [1] to [6], the sense refers to the five senses or a sense that is further subdivided from the five senses.

[0022] [8] A material recommendation method comprises: an input step for receiving input of information indicating a product category and an emotional language, wherein the emotional language indicates a language for expressing emotions; a material parameter extraction step for extracting the value of the sensory parameter of each material corresponding to the product category input in the input step based on establishing corresponding material parameter information between the material used in the product belonging to the product category and the value of the sensory parameter indicating the degree of feeling felt by a person with respect to the material; an emotional language parameter extraction step for extracting the value of the sensory parameter corresponding to the emotional language input in the input step based on establishing corresponding emotional language parameter information between the emotional language and the value of the sensory parameter; and a recommended material determination step for determining the material to be recommended for the emotional language input in the input step based on the value of the sensory parameter of each material extracted in the material parameter extraction step and the value of the sensory parameter extracted in the emotional language parameter extraction step.

[0023] [9] A program for causing a computer to execute: an input step for receiving input of information representing a product category and an emotional language, wherein the emotional language represents a language for expressing emotions; a material parameter extraction step for extracting the value of the sensory parameter of each material corresponding to the product category input in the input step based on establishing corresponding material parameter information between the material used in the product belonging to the product category and the value of the sensory parameter representing the degree of feeling felt by a person with respect to the material; an emotional language parameter extraction step for extracting the value of the sensory parameter corresponding to the emotional language input in the input step based on establishing corresponding emotional language parameter information between the emotional language and the value of the sensory parameter; and a recommended material determination step for determining the material recommended for the emotional language input in the input step based on the value of the sensory parameter of each material extracted in the material parameter extraction step and the value of the sensory parameter extracted in the emotional language parameter extraction step.

[0024] Effects of the Invention

[0025] According to the present invention, materials can be recommended according to the emotions of a person. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 : is a block diagram showing the functional structure of the material recommendation device 100 according to one embodiment of the present disclosure.

[0027] Figure 2 1 is a diagram showing an example of emotional language input to the material recommendation device 100 according to one embodiment of the present disclosure.

[0028] Figure 3 : is a diagram showing an example of five sense parameters used in the material recommendation device 100 according to one embodiment of the present disclosure.

[0029] Figure 4 1 is a diagram showing an example of material parameter information 111 stored in the parameter information storage unit 110 of the material recommendation device 100 according to one embodiment of the present disclosure.

[0030] Figure 5 1 is a diagram showing an example of the emotional language parameter information 112 stored in the parameter information storage unit 110 of the material recommendation device 100 according to the embodiment of the present disclosure.

[0031] Figure 6 1 is a diagram showing an example of the modality correction information 113 stored in the parameter information storage unit 110 of the material recommendation device 100 according to the embodiment of the present disclosure.

[0032] Figure 71 is a diagram showing an example of the modality correction information 113 stored in the parameter information storage unit 110 of the material recommendation device 100 according to the embodiment of the present disclosure.

[0033] Figure 8 1 is a flowchart showing the operation of the material recommendation device 100 according to one embodiment of the present disclosure.

[0034] Fig. 9 1 is a diagram showing an example of a correction process of a material parameter value performed by the modality correction unit 133 of the material recommendation device 100 according to one embodiment of the present disclosure.

[0035] Fig.10 1 is a diagram showing an example of a correction process of the value of the emotional language parameter performed by the modality correction unit 133 of the material recommendation device 100 according to one embodiment of the present disclosure.

[0036] Fig.11 1 is a diagram showing an example of a calculation process of the distance between coordinates performed by the recommended material determination unit 134 of the material recommendation device 100 according to one embodiment of the present disclosure.

[0037] Fig.12 1 is a diagram showing an example of information output by the output unit 140 of the material recommendation device 100 according to one embodiment of the present disclosure.

[0038] Fig.13 2 is a diagram showing another example of information output by the output unit 140 of the material recommendation device 100 according to the embodiment of the present disclosure. DETAILED DESCRIPTION

[0039] Hereinafter, one embodiment of the present disclosure will be described with reference to the drawings.

[0040] [Structure of material recommendation device]

[0041] Hereinafter, the structure of the material recommendation device 100 according to the embodiment will be described. Figure 1 : is a block diagram showing the functional structure of the material recommendation device 100 according to one embodiment of the present disclosure.

[0042] The material recommendation device 100 is, for example, an information processing device configured using a general-purpose computer. The material recommendation device 100 receives input of language (hereinafter referred to as "emotional language") that expresses the user's desired feeling for the material used in the product and that expresses the user's emotions. The material recommendation device 100 determines at least one recommended material based on the input emotional language, and outputs information indicating the determined material.

[0043] Figure 1 1 is a block diagram showing the functional structure of the material recommendation device 100 in the first embodiment of the present disclosure. Figure 1 As shown, the material recommendation device 100 includes a parameter information storage unit 110 , an input unit 120 , a material parameter extraction unit 131 , an emotional language parameter extraction unit 132 , a modality correction unit 133 , a recommended material determination unit 134 , and an output unit 140 .

[0044] The parameter information storage unit 110 is composed of a computer-readable storage medium. The computer-readable storage medium mentioned here is, for example, a magnetic hard disk device or a semiconductor storage device. The parameter information storage unit 110 stores various parameter information used to determine the recommended material in advance. Figure 1 As shown, the parameter information storage unit 110 stores at least material parameter information 111, emotional language parameter information 112, and modality correction information 113. Details of each parameter information will be described in detail later.

[0045] The input unit 120 is composed of an input device that receives input operations from the user. The so-called input device here is, for example, a keyboard, a mouse, a touch panel, and an input button. Figure 1 As shown, the input unit 120 includes a product category input unit 121 , an emotional language input unit 122 , and a usage status input unit 123 .

[0046] The product category input unit 121 obtains information indicating the product category input by the user (hereinafter referred to as "product category information"). The product category input unit 121 outputs the obtained product category information to the material parameter extraction unit 131. The product category information mentioned here is information indicating the category of materials used in the product, such as "food packaging materials", "housing building materials", and "artificial leather materials for bags".

[0047] The emotional language input unit 122 obtains information indicating the emotional language input by the user (hereinafter referred to as "emotional language"). The emotional language input unit 122 outputs the obtained emotional language to the emotional language parameter extraction unit 132. The so-called emotional language here refers to sentences such as adjectives used in general sensory evaluation to quantify human feelings. Such adjectives include sentences that cannot be captured only in specific feelings among the five senses, such as "favorite", "strange", "comfortable" and "high-end".

[0048] In addition, in general sensory evaluation, when evaluating people's emotions by numericalizing them, evaluation based on the SD (Semantic Differential) method is often used. The so-called SD method is a method that uses pairs of adjectives with opposite meanings to measure people's impressions (sensitivity) of objects and evaluate them numerically. The SD method is characterized by the structure that can express impressions with a small number of factors. For example, in the SD method, a structure consisting of three basic factors interpreted as evaluative factors, activity factors, and effectiveness factors is found. This structure is called EPA (Evaluation, Potency, Activity) structure.

[0049] The evaluation factor is represented by pairs of adjectives that express comprehensive evaluations related to the value of the object, such as "like-hate", "good-bad", and "beautiful-ugly". In addition, the evaluation factor varies depending on the type of object being evaluated (shape, color, sound, etc.), and is also composed of pairs of adjectives that express the nature of the object, such as "stable-unstable", "clear-turbid", and "warm-cold".

[0050] Activity and Potency are both composed of expressions about the properties of the object. Activity is expressed by pairs of adjectives such as "loud-quiet", "dynamic-static", and "flashy-plain". Potency is expressed by pairs of adjectives such as "hard-soft", "sharp-blunt", and "tense-relaxed".

[0051] Figure 2 1 is a diagram showing an example of emotional language input to the material recommendation device 100 in one embodiment of the present disclosure. The evaluative factor and the active factor among the three basic factors mentioned above contain many adjectives that directly express people's emotions. In this embodiment, as an example, sentences of the evaluative factor and the active factor among the factors of the EPA structure defined in the impression evaluation based on the SD method are defined as emotional language. In addition, instead of defining a pair of adjectives as emotional language, each side of the paired adjectives is defined as one emotional language.

[0052] The usage status input unit 123 acquires information indicating the usage status of the product input by the user (hereinafter referred to as “usage status information”). The usage status input unit 123 outputs the acquired usage status information to the modal correction unit 133 .

[0053] The so-called usage status information mentioned here is information that can determine which sense among the five senses the user uses to feel the impression from the product. In addition, the so-called usage status information is information that can determine in what order the user feels the impression based on each sense when the user feels the impression through multiple senses among the five senses. For example, the so-called usage status information is information that can determine whether the user sees the product first and then touches the product, or touches the product first and then sees the product, when the user feels the impression through two senses of the five senses, namely, vision and touch.

[0054] The material parameter extraction unit 131, the emotional language parameter extraction unit 132, the modality correction unit 133, and the recommended material determination unit 134 are functional units implemented by, for example, a CPU (Central Processing Unit) or other processor reading and executing a program pre-stored in a storage unit (not shown). In addition, the above-mentioned storage unit and parameter information storage unit 110 may also be configured using the same storage medium.

[0055] The material parameter extraction unit 131 acquires the product category information output from the product category input unit 121. The material parameter extraction unit 131 also refers to the material parameter information 111 stored in the parameter information storage unit 110. The material parameter extraction unit 131 extracts the value of the material parameter associated with the product category corresponding to the acquired product category information from the material parameter information 111.

[0056] The material parameter information 111 is information indicating the degree of correlation between the material and the five sense parameters. In addition, the so-called five sense parameters mentioned here are composed of items indicating the five senses or items indicating the senses further subdivided from each of the five senses. In addition, in this embodiment, the five sense parameters are composed of items indicating the senses further subdivided from each of the five senses.

[0057] Figure 3 is a diagram showing an example of five sense parameters used in the material recommendation device 100 according to one embodiment of the present disclosure. Figure 3 As shown, the sense of touch is further classified into five sense parameters such as "hardness", "roughness", "friction" and "warmth and coldness". In addition, the sense of vision is further classified into five sense parameters such as "color 1", "color 2", "color 3", "texture 1", "texture 2" and "texture 3".

[0058] In addition, regarding the sense of touch, it is known that it can be classified into four elements: hardness, roughness, friction, and temperature. In addition, specifically, for example, as the five sense parameters of the "roughness" of the sense of touch, the values ​​of the average roughness Ra and the maximum height Rz specified by the standard (JIS B 0601) can be used. Alternatively, parameters and values ​​whose dimensions are reduced by principal component analysis or factor analysis of these average roughness Ra and maximum height Rz can be used. In addition, for example, as the five sense parameters of the "hardness" of the sense of touch, the value of the Young's modulus calculated based on the force and displacement of the indentation can be used. In addition, for example, as the five sense parameters of the "friction" of the sense of touch, the value of the dynamic friction coefficient when touching the material can be used.

[0059] In addition, when the material recommendation device 100 is used for various materials, it is desirable to obtain a parameter value uniformly evaluated for all materials using a wearable device, etc. By equipping a wearable device with a human hand and actually touching the material, friction and hardness can be uniformly measured.

[0060] In addition, regarding vision, Figure 3 The five sense parameters of vision shown include "color 1", "color 2" and "color 3", but it is generally known that color can be expressed by three variables such as L*a*b* or L*C*H. For example, as the five sense parameters of "color 1", "color 2" and "color 3", these three variables can be used respectively. In addition, the material recommendation device 100 can use any three variables of any color space, but it is necessary to use the same three variables of the color space for all materials for evaluation.

[0061] In addition, regarding vision, it is not known that it can be clearly classified into a limited number of elements like touch. However, it has recently been learned that the elements of color and texture are important factors that affect the visual impression of people. Here, the so-called texture is physical property information related to reflectivity, represented by the bidirectional reflectance distribution function BRDF. BRDF (Bidirectional Reflectance Distribution Function) is the angular distribution characteristics of reflected light when light is incident from a certain angle. Generally speaking, in CG (Computer Graphics), the texture of the material is reconstructed by rendering using the parameters of the BRDF. Of course, for these parameters, parameters and values ​​with reduced dimensions can also be used by performing principal component analysis or factor analysis.

[0062] In addition, regarding the sense of taste, the type and amount of taste chemical components are considered as factors that affect the impression. In addition, regarding the sense of smell, the type and concentration of olfactory chemical components are considered as factors that affect the impression. In addition, regarding the sense of hearing, the signal intensity and power spectrum are considered as factors that affect the impression. Of course, for these parameters, the parameters and values ​​with reduced dimensions can also be used by performing principal component analysis or factor analysis.

[0063] In addition, the value of the material parameter may be, for example, an actual measurement value obtained by a certain measurement method, or a value obtained by standardizing the values ​​of the material parameter of all materials. In the case of using the actual measurement value itself, it is necessary to unify the units of the values ​​of the material parameter of all materials in advance. For example, when the value of the average roughness Ra specified by the specification is used as the material parameter of "roughness", the values ​​of the material parameter of "roughness" of all materials may be unified into values ​​in units such as millimeters or micrometers.

[0064] In addition, in the impressions of materials used in industrial products such as food packaging materials, housing building materials, and artificial leather materials for bags, it is believed that the impressions of vision and touch, in particular, account for a large proportion among the five senses.

[0065] Figure 4 1 is a diagram showing an example of material parameter information 111 stored in the parameter information storage unit 110 of the material recommendation device 100 according to one embodiment of the present disclosure. The material parameter information 111 is data showing the correlation between the material and the five sense parameters described above. Hereinafter, the five sense parameters associated with the material are referred to as "material parameters".

[0066] exist Figure 4 The material parameter information 111 shown here associates the product category, the product number, the information indicating the material structure of the product, and the material parameter with each other. Each row of the material parameter information 111 is information related to one material.

[0067] For example, in Figure 4 In the material parameter information 111 shown, in the first row, the value "food packaging material" is registered in the item "product category", and the value "101a001b001" is set in the item "product number". This means that the material assigned with the product number "101a001b001" is a material used in food packaging materials. Figure 4In the material parameter information 111 shown, in the first row, a value of "101" is set in the item "Main structure" of the item "Material structure of product", a value of "a001" is set in the item "Additive 1", and a value of "b001" is set in the item "Additive 2". This means that the material assigned the product number "101a001b001" is a material including the main structure assigned the identification information "101", the additive assigned the identification information "a001", and the additive assigned the identification information "b001".

[0068] In addition, each material parameter in the material parameter information 111 is registered with a value normalized by, for example, 0. Normalization is an operation of subtracting the average value of the entire database of each material parameter from the value of each material parameter and dividing it by the standard deviation, so that the average is 0 and the variance is 1. In this way, each material is digitized by the five sense parameters (material parameters). Figure 4 For example, the value of the five sense parameter "hardness" of the product number "101a001b001" is "0.12", and the value of the five sense parameter "warmth" is "0.03".

[0069] In addition, depending on the product category, there are materials that do not affect the impression of some of the five senses (such as hearing and taste). For example, taste may have a slight effect on the impression of food packaging materials, but it has no effect on the impression of residential building materials or artificial leather materials for bags.

[0070] The emotional language parameter extraction unit 132 obtains the emotional language output from the emotional language input unit 122. Figure 2 As an example in FIG. 1 , the information language is a sentence expressing a person's emotion. In addition, the emotion language parameter extraction unit 132 refers to the emotion language parameter information 112 stored in the parameter information storage unit 110. The emotion language parameter extraction unit 132 extracts the value of the emotion language parameter corresponding to the acquired emotion language from the emotion language parameter information 112.

[0071] The emotional language parameter information 112 is information indicating the degree of correlation between emotional language and five sense parameters. Figure 3 Each of the five senses is further subdivided into items of sense as shown.

[0072] Figure 51 is a diagram showing an example of the emotional language parameter information 112 stored in the parameter information storage unit 110 of the material recommendation device 100 in one embodiment of the present disclosure. The emotional language parameter information 112 is data showing the correlation between the emotional language and the five sense parameters described above. Hereinafter, the five sense parameters corresponding to the emotional language are referred to as "emotional language parameters".

[0073] exist Figure 5 In the emotional language parameter information 112 shown, the emotional language and the emotional language parameter are associated with each other. Each row of the emotional language parameter information 112 is information related to one emotional language. Figure 5 For example, the value of the five sense parameter "texture 1" of the emotional language "beautiful" is "0.73", and the value of the five sense parameter "texture 2" is "-0.15".

[0074] In this way, each emotional language is digitized in advance using the five sense parameters (emotional language parameters). In addition, as a method of digitization, for example, the following method is considered: for each material, a plurality of subjects are evaluated in advance using emotional language, and a regression model of emotional language and five sense parameters is created to determine the numerical value. In addition, these emotional language parameters can also use standardized values ​​using the entire emotional language database in the same way as the material parameters.

[0075] and Figure 4 Similarly, in each emotional language parameter of the emotional language parameter information 112, a normalized value based on 0 is registered. Normalization is an operation of subtracting the average value of the entire database of each emotional language parameter from the value of each emotional language parameter and dividing it by the standard deviation, and setting the standard deviation to an average of 0 and a variance of 1. In this way, each emotional language is digitized through the five sense parameters (emotional language parameters).

[0076] The modal correction unit 133 obtains the usage status information output from the usage status input unit 123. As described above, the usage status information is information that specifies which sense among the five senses the user follows and how the user feels the impression. In addition, the modal correction unit 133 refers to the modal correction information 113 stored in the parameter information storage unit 110. The modal correction unit 133 corrects the values ​​of the material parameters extracted by the material parameter extraction unit 131 and the values ​​of the emotional language parameters extracted by the emotional language parameter extraction unit 132, respectively, based on the obtained usage status information and the modal correction information 113.

[0077] Specifically, for example, the modal correction unit 133 determines, based on the usage status information, the order in which the user perceives the impressions based on the senses when the user perceives the impressions through multiple senses among the five senses. The modal correction unit 133 extracts a coefficient group for considering the cross-modal effect generated in the determined usage status from the modal correction information 113. The modal correction unit 133 multiplies the extracted coefficient group by the value of the material parameter extracted by the material parameter extraction unit 131 and the value of the emotional language parameter extracted by the emotional language parameter extraction unit 132.

[0078] Generally speaking, the so-called cross-modal effect means that the impressions respectively perceived by two of the five senses, such as vision and touch, are related to each other. In addition, generally speaking, the so-called multimodal effect means that the impressions respectively perceived by multiple senses, such as vision, touch and hearing, are related to each other. However, the cross-modal effect mentioned in this embodiment also includes the meaning of multimodal effect.

[0079] For example, when a user recognizes a bag, he or she may carefully look at the bag visually and then touch the bag to confirm the feel. On the other hand, the user may also check the bag visually after touching the bag to confirm the feel. In addition, the user's comprehensive impression of the material may be different depending on which of the visual and tactile senses is felt first. For example, a material such as artificial leather for a bag produces a cross-modal effect, and gives the user a different impression depending on the order in which multiple senses work.

[0080] The material recommendation device 100 in this embodiment corresponds to considering the cross-modal effect based on the usage status information obtained. Specifically, as described above, the modal correction unit 133 multiplies the coefficient group for considering the cross-modal effect by the value of the material parameter extracted by the material parameter extraction unit 131 and the value of the emotional language parameter extracted by the emotional language parameter extraction unit 132. Thus, the value of the material parameter and the value of the emotional language parameter are corrected to reflect the influence of the cross-modal effect corresponding to the usage status information.

[0081] In the present embodiment, the modal correction information 113 (for example, Figure 6 The modality correction information 113 shown in FIG. 1 ), and the modality correction information 113 including the coefficient group to be multiplied by the value of the emotional language parameter (for example Figure 7 The modal correction information 113 shown is respectively stored in advance in the parameter information storage unit 110 .

[0082] Figure 6 as well as Figure 71 is a diagram showing an example of the modality correction information 113 stored in the parameter information storage unit 110 of the material recommendation device 100 according to the embodiment of the present disclosure.

[0083] exist Figure 6 2 shows an example of the modal correction information 113 including a coefficient group to be multiplied by the value of the material parameter. Figure 6 The coefficient group corresponding to the usage status information used in the correction of the material parameter value is shown. Figure 6 As shown, in the case of the coefficient multiplied by the value of the material parameter, for example, the coefficient value of the feeling felt first in the cross-modality is preset to 1, the coefficient value of the feeling felt later is preset to a value greater than 0 and less than 1, and the coefficient values ​​of other feelings are preset to 0.

[0084] In addition, similar to the case where the aforementioned emotional language is pre-digitized using five sense parameters, as a method of digitizing such a coefficient, for example, the following method can be considered: for each usage condition, a sensory evaluation using each material is performed in advance on multiple subjects, and a regression model of the usage condition and the material is created to determine the numerical value.

[0085] exist Figure 7 , an example of the modality correction information 113 including a coefficient group to be multiplied by the value of the emotional language parameter is shown. Figure 7 The coefficient group corresponding to the usage status information is used to correct the value of the emotional language parameter. Figure 7 As shown, in the case of the coefficient to be multiplied by the value of the emotional language parameter, for example, 1 is preset for the coefficient value of the sense that produces the cross-modal effect, and 0 is preset for the coefficient value of the other senses.

[0086] The material parameter extraction unit 131 outputs information indicating the material parameter corrected by the modality correction unit 133 to the recommended material determination unit 134 . The emotional language parameter extraction unit 132 outputs information indicating the emotional language parameter corrected by the modality correction unit 133 to the recommended material determination unit 134 .

[0087] The recommended material determination unit 134 obtains the information indicating the material parameter output from the material parameter extraction unit 131 and the information indicating the emotional language parameter output from the emotional language parameter extraction unit 132. The recommended material determination unit 134 calculates the distance between the coordinates of the material parameter value and the emotional language parameter value for each material based on the obtained information. The recommended material determination unit 134 determines the recommended material based on the calculated distance. For example, the recommended material determination unit 134 determines the material with the shortest calculated distance or a given number of materials in order from the side with the shortest calculated distance as the recommended material.

[0088] The recommended material determination unit 134 outputs information indicating the recommended material to the output unit 140. In addition, the recommended material determination unit 134 may also output information indicating the distance between the value of the material parameter and the value of the emotional language parameter for each recommended material.

[0089] In addition, here, when selecting materials that match the emotional language, a method of simply comparing the distance between the material parameter value and the emotional parameter value is used. As other methods, for example, a method of solving an inverse problem such as Bayesian optimization can be used, and a method of solving a direct problem by performing a grid search to minimize the error can also be used. In addition, it is also possible to use an AutoML (Automated Machine Learning) such as DataRobot (registered trademark) to perform the above-mentioned operations on behalf of the user.

[0090] The output unit 140 is composed of an output device that presents information to the user. The so-called output device here is, for example, a display device such as a liquid crystal display (LCD), a CRT (Cathode Ray Tube), or an organic EL (Electro-Luminescence) display. In addition, the output unit 140 can also be a communication interface that sends the information to an external device that presents information to the user. The so-called external device is, for example, an information processing device such as a general-purpose computer, various displays or printers, etc.

[0091] The output unit 140 acquires the information indicating the recommended material output from the recommended material determination unit 134. The output unit 140 presents the information indicating the recommended material to the user.

[0092] [Operation of the material recommendation device]

[0093] Hereinafter, an example of the operation of the material recommendation device 100 of the present disclosure will be described with reference to a specific example. Figure 8 1 is a flowchart showing the operation of the material recommendation device 100 according to one embodiment of the present disclosure.

[0094] First, the product category input unit 121 receives the user's input of product category information. For example, the product category information is displayed on a display device such as a monitor. Figure 4 The leftmost column of the material parameter information 111 shown is a list of values ​​of the item "product category" ("food packaging materials", "housing building materials", "artificial leather materials for bags", etc.). The user performs an input operation to select a desired product category from the displayed list using an input device such as a mouse. Here, as an example, the user selects the product category of "artificial leather materials for bags".

[0095] The product category input unit 121 outputs product category information indicating “artificial leather material for bags” to the material parameter extraction unit 131. The material parameter extraction unit 131 acquires the product category information output from the product category input unit 121 (step S01).

[0096] Next, the emotional language input unit 122 receives the user's emotional language input. For example, the emotional language is displayed on a display device such as a monitor. Figure 2 A list of emotional language is shown ("favorite", "gorgeous", "loving", etc.). The user performs an input operation to select the desired emotional language from the displayed list using an input device such as a mouse. In addition, here, as an example, the user selects the emotional language "beautiful". In addition, it is not limited to this example, and multiple emotional languages ​​can also be selected. In this case, the following processing is performed on each emotional language.

[0097] The emotional language input unit 122 outputs the emotional language "beautiful" to the emotional language parameter extraction unit 132. The emotional language parameter extraction unit 132 acquires the emotional language output from the emotional language input unit 122 (step S02).

[0098] Next, the usage status input unit 123 receives input of the user's usage status information. Figure 6 or Figure 7 The second column from the left of the modal correction information 113 shown is a list of values ​​of the "first" and "after" items of the "cross-modal effect" ("tactile" "none", "visual" "none", "auditory" "none", etc.). The user imagines the usage status when using the product, and performs an input operation of selecting the desired usage status information from the displayed list, for example, using an input device such as a mouse. In addition, here, as an example, the user selects the usage status information in which the first sense felt is "visual" and the second sense felt is "tactile".

[0099] The usage status input unit 123 outputs usage status information indicating that the first sense felt is “vision” and the second sense felt is “tactile” to the modality correction unit 133. The modality correction unit 133 acquires the usage status information output from the usage status input unit 123 (step S03).

[0100] In addition, a list of options consisting of phrases such as "determine whether to buy by just looking at the bag" and "determine whether to buy by checking the touch feeling after looking at the bag" is displayed on a display device such as a monitor, and the user can select from the list. In this case, for example, the parameter information storage unit 110 pre-stores information that associates these options with one or more senses. That is, for example, information that associates "determine whether to buy by just looking at the bag" with "sight", and information that associates "determine whether to buy by checking the touch feeling after looking at the bag" with "sight" and "tactile sense" is pre-stored.

[0101] Then, when the modal correction unit 133 obtains information indicating that the user has selected an option such as "confirm the touch sensation after viewing the bag and decide whether to purchase it", for example, from the usage status input unit 123, the modal correction unit 133 can determine the usage status information in which the sense to be felt first is "visual" and the sense to be felt later is "tactile" by referring to the above information pre-stored in the parameter information storage unit 110.

[0102] Next, the material parameter extraction unit 131 extracts the value of the material parameter corresponding to the product category corresponding to the acquired product category information from the material parameter information 111 pre-stored in the parameter information storage unit 110 (step S04). For example, here, after the user selects the product category of "artificial leather material for bags", the material parameter extraction unit 131 extracts all the material parameters. Figure 4 The value of the item “Product Category” of the material parameter information 111 shown is data (value of material parameter) of the row “Artificial Leather Material for Bags”.

[0103] In addition, Figure 4 In the material parameter information 111 shown, there are five rows (m=5) in which the value of the item "Product Category" is "Artificial Leather Material for Bags". That is, five materials (product numbers "301a201b101" to "305a201b105") are registered as artificial leather materials for bags. If the number of material parameters (sensory parameters) corresponding to each material is n, the material parameter extraction unit 131 extracts m (five) matrices consisting of 1 row and n columns (hereinafter referred to as "matrix A").

[0104] Next, the modal correction unit 133 corrects the value of the material parameter extracted by the material parameter extraction unit 131 based on the acquired usage status information and the modal correction information 113 (step S05 ).

[0105] For example, here, the user selects the usage status information that the first sense felt is "vision" and the second sense felt is "tactile". Figure 6The modal correction unit 133 shows the data (coefficient group) of the row whose value of the "first" item of the "cross-modal effect" item is "visual" and whose value of the "last" item is "tactile". If the number of material parameters corresponding to each material is set to n, the modal correction unit 133 extracts a matrix consisting of 1 row and n columns. Then, the modal correction unit 133 generates a diagonal matrix consisting of n rows and n columns (hereinafter referred to as "diagonal matrix B") with the extracted matrix as diagonal components.

[0106] The modal correction unit 133 corrects the value of the material parameter by multiplying each of the extracted matrices A by the generated diagonal matrix B. Thus, the value of the material parameter is corrected to a value that takes into account the cross-modal effect corresponding to the selected usage status information.

[0107] Fig. 9 1 is a diagram showing an example of a correction process of a material parameter value performed by the modal correction unit 133 of the material recommendation device 100 according to one embodiment of the present disclosure. Fig. 9 As an example, the matrix calculation when correcting the value of the material parameter of the artificial leather material for a bag assigned a product number such as "301a201b101" is shown. Fig. 9 As shown, by Figure 4 The material parameter information 111 shown in FIG. 1 is used to extract the material parameter values, and the matrix A is formed by the material parameter values ​​extracted from Figure 6 The matrix composed of the values ​​of the corrected material parameters ("-0.09", "0.71", "0.69", "0.26", ...) is calculated by multiplying the diagonal matrix B generated by the coefficient group extracted from the modal correction information 113 shown.

[0108] Next, the emotional language parameter extraction unit 132 extracts the value of the emotional language parameter corresponding to the acquired emotional language from the emotional language parameter information 112 pre-stored in the parameter information storage unit 110 (step S06). For example, here, since the user selects the emotional language "beautiful", the emotional language parameter extraction unit 132 extracts Figure 5 The data (value of the emotional language parameter) of the row in which the value of the item “emotional language” of the emotional language parameter information 112 is “beautiful” is shown.

[0109] If the number of emotional language parameters (sensation parameters) corresponding to each emotional language is set to n, the emotional language parameter extraction unit 132 extracts m matrices consisting of 1 row and n columns (hereinafter referred to as "matrices C"). In addition, the value of m here represents the number of emotional languages ​​selected by the user. In addition, as an example, since the user selects only one emotional language ("beautiful"), the emotional language parameter extraction unit 132 extracts one matrix C.

[0110] Next, the modality correction unit 133 corrects the value of the emotional language parameter extracted by the emotional language parameter extraction unit 132 based on the acquired usage status information and the modality correction information 113 (step S07 ).

[0111] For example, here, the user selects the usage status information that the first sense felt is "vision" and the second sense felt is "tactile". Figure 7 The modal correction information 113 of the "cross-modal effect" item shows a row of data (coefficient group) in which the value of the "first" item is "visual" and the value of the "last" item is "tactile". If the number of emotional language parameters corresponding to each emotional language is set to n, the modal correction unit 133 extracts a matrix consisting of 1 row and n columns. Then, the modal correction unit 133 generates a diagonal matrix consisting of n rows and n columns (hereinafter referred to as "diagonal matrix D") with the extracted matrix as a diagonal component.

[0112] The modality correction unit 133 corrects the value of the emotional language parameter by multiplying each extracted matrix C (here, one) by the generated diagonal matrix D. In this way, the value of the emotional language parameter is corrected to a value that takes the cross-modality effect into consideration.

[0113] Fig.10 1 is a diagram showing an example of a correction process of the value of the emotional language parameter performed by the modality correction unit 133 of the material recommendation device 100 according to one embodiment of the present disclosure. Fig.10 Matrix calculations when correcting the values ​​of the emotional language parameters corresponding to the emotional language "beautiful" are shown in FIG. Fig.10 As shown, by Figure 5 The matrix C consisting of the values ​​of the emotional language parameters extracted from the material parameter information 111 shown in FIG. Figure 7 The matrix composed of the values ​​of the corrected emotional language parameters ("0.03", "-0.81", "-0.62", "0.03", ...) is calculated by multiplying the diagonal matrix D generated by the coefficient group extracted from the modality correction information 113 shown.

[0114] The material parameter extraction unit 131 outputs the corrected value of the material parameter to the recommended material determination unit 134. In addition, the emotion-language parameter extraction unit 132 outputs the corrected value of the emotion-language parameter to the recommended material determination unit 134.

[0115] Next, the recommended material determination unit 134 obtains the value of the corrected material parameter output from the material parameter extraction unit 131 and the value of the corrected emotional language parameter output from the emotional language parameter extraction unit 132. The recommended material determination unit 134 regards the obtained matrix of material parameters and the matrix of emotional language parameters as coordinates, and calculates the distance between the coordinates for each material. The recommended material determination unit 134 determines the recommended material based on the calculated distance (step S08). For example, the material whose coordinates of the emotional language parameter corresponding to the emotional language specified by the user and the coordinates of the material parameter corresponding to the material are close can be regarded as having a strong relationship with the emotional language, and can be determined in sequence from a plurality of materials.

[0116] Fig.11 1 is a diagram showing an example of a calculation process of the distance between coordinates performed by the recommended material determination unit 134 of the material recommendation device 100 according to one embodiment of the present disclosure. Fig.11 The following processing is shown in Fig.10 The matrix consisting of the values ​​of the corrected emotional language parameters calculated by the correction processing shown in FIG. 1 is subtracted from the matrix consisting of the values ​​of the corrected emotional language parameters calculated by Fig. 9 The distance between the coordinates is calculated by squaring the respective values ​​of the calculated matrix components composed of the values ​​of the corrected material parameters calculated by the correction processing shown, and further summing up the calculated values ​​and performing root calculation.

[0117] For example, in Fig.11 In the example, since the value of the emotional language parameter of "hardness" is "0.03" and the value of the material parameter of "hardness" is "-0.09", if a subtraction operation is performed and the square is taken, it becomes "0.0144". The recommended material determination unit 134 calculates the value obtained by summing up the component values ​​calculated for each of the five sense parameters and then performing a root calculation as the distance between the coordinates of the emotional language parameter and the material parameter. In addition, the distance between the coordinates calculated here is, for example, the square root of "4.0938+α". The calculation of the distance between the coordinates is not limited to this example, and can also be calculated by other methods suppressed in the technical field.

[0118] The recommended material determination unit 134 performs the above calculation for each material and each emotional language, and calculates the distance between the coordinates. The recommended material determination unit 134 determines the recommended material based on the calculated distance between the coordinates. The recommended material determination unit 134 outputs information indicating the recommended material and information indicating the distance between the coordinates calculated for the material to the output unit 140.

[0119] The output unit 140 obtains information indicating the recommended material and information indicating the distance between coordinates calculated for the material output from the recommended material determination unit, and presents the information indicating the recommended material and information indicating the distance between coordinates calculated for the material to the user (step S09 ).

[0120] Fig.12 is a diagram showing an example of information output by the output unit 140 of the material recommendation device 100 according to one embodiment of the present disclosure. Fig.12 As shown in FIG. 1 , the information output by the output unit 140 includes, for example, the selected product category (“artificial leather material for bags”), the selected emotional language (“beautiful”), and the value of the distance between the coordinates of each material (for each product number). In this case, the output unit 140 can display the more recommended materials at the top by outputting them in order from the materials with the shorter distance between the coordinates. In addition, as Fig.12 As shown, by displaying the distance between coordinates in a bar graph, the user can more intuitively recognize the recommendation level of each material. However, the present invention is not limited to this example, and the distance between coordinates may be displayed in a given order of product numbers.

[0121] In addition, an example of a method of presenting recommended materials by the output unit 140 when a user selects a plurality of emotion words is given below. Fig.13 1 is a diagram showing another example of information output by the output unit 140 of the material recommendation device 100 according to one embodiment of the present disclosure. Here, the user selects an emotional language such as "flashy" and an emotional language such as "manly".

[0122] exist Fig.13 In the example of , the information output by the output unit 140 includes, for example, the selected product category ("artificial leather material for bags"). In addition, the information output by the output unit 140 includes, for example, the value of the distance between coordinates of each material (for each product number) for each of the selected emotion languages ​​("flashy" and "manly"). Furthermore, the information output by the output unit 140 includes, for example, the average value of the distance between coordinates in a plurality of emotion languages ​​for each material.

[0123] By presenting information in this way by the output unit 140, the user can recognize, for example, that the artificial leather material for a bag that is closest to the impression of "flashy" is the material with a coordinate distance of "8.7" and a product number of "303a201b102". In addition, the user can recognize, for example, that the artificial leather material for a bag that is closest to the impression of "manly" is the material with a coordinate distance of "8.3" and a product number of "303a201b104". In addition, the user can recognize, for example, that the artificial leather material for a bag that has both the impression of "flashy" and the impression of "manly" at a high average is the material with a coordinate distance of "9.4" and a product number of "303a201b104". When more than three emotion languages ​​are specified, the average value of the three emotion languages ​​may be output, or the average value may be output for every combination of two of the three emotion languages.

[0124] Thus, the user can select, for example, a material having the shortest distance between all coordinates, or can select a material having an impression that is closest on average to a plurality of selected emotional words.

[0125] In addition, for example, the output unit 140 can generate and prompt a plot that can visually grasp the coordinates of the material parameters and the coordinates of the emotional language parameters. In addition, for example, the output unit 140 can also perform principal component analysis and multidimensional scaling analysis on the coordinates of the material parameters and the coordinates of the emotional language parameters, and visually prompt the analysis results. For example, in the case of principal component analysis, the output unit 140 can use the first principal component and the second principal component as two axes to prompt the positional relationship between the coordinates of the material parameters and the emotional language parameters in a two-dimensional graph based on BiPlot.

[0126] As described above, the material recommendation device 100 in one embodiment of the present disclosure includes the input unit 120, the material parameter extraction unit 131, the emotional language parameter extraction unit 132, and the recommended material determination unit 134. The input unit 120 receives input of information indicating a product category and emotional language indicating a language expressing emotions. The material parameter extraction unit 131 extracts the value of the sensory parameter (material parameter) of each material corresponding to the product category input to the input unit 120 based on the material parameter information that associates the material used in the product belonging to the product category with the value of the sensory parameter indicating the degree of feeling felt by a person with respect to the material. The emotional language parameter extraction unit 132 extracts the value of the sensory parameter (emotional language parameter) corresponding to the emotional language input to the input unit 120 based on the emotional language parameter information that associates the emotional language with the value of the sensory parameter. The recommended material determination unit 134 determines the material recommended for the emotional language input to the input unit 120 based on the value of the sensory parameter of each material extracted by the material parameter extraction unit 131 and the value of the sensory parameter extracted by the emotional language parameter extraction unit 132.

[0127] By having such a configuration, the material recommendation device 100 can automatically recommend appropriate materials according to the emotions of a person, for example, without relying on the senses of a specific person in charge of recommending the material.

[0128] In addition, the material recommendation device 100 corrects the value of the sensory parameter for each material and the value of the sensory parameter corresponding to the emotional language in order to consider the influence of the cross-modal effect. With such a structure, the material recommendation device 100 can recommend appropriate materials according to the emotions of a person with higher accuracy.

[0129] (Variation)

[0130] It can also be omitted Figure 1 The material recommendation device 100 shown in the figure has a simpler structure of the modal correction information 113, the usage status input unit 123, and the modal correction unit 133 among the functional structures of the material recommendation device 100. That is, it is also possible to set it as a material recommendation device that omits the structure of correcting the five sense parameters in consideration of the cross-modal effect among the functions of the material recommendation device 100. The material recommendation device 100 can also correct the value of the sensory parameter of each material without corresponding to the usage status information, and determine the recommended material based on the value of the sensory parameter. In this way, by setting it to a simpler structure, the device cost, operation cost, and man-hours for preparing the modal correction information 113 of the material recommendation device 100 can be reduced.

[0131] In addition, the program for realizing the function of the material recommendation device 100 in the embodiment can also be recorded in a computer-readable recording medium, so that the computer system reads and executes the program recorded in the recording medium, thereby performing the above-mentioned processing. In addition, the so-called "computer system" mentioned here includes hardware such as OS and peripheral devices. In addition, the "computer system" also includes a WWW system with a homepage providing environment (or display environment). In addition, the so-called "computer-readable recording medium" refers to a removable medium such as a floppy disk, a magneto-optical disk, a ROM, a CD-ROM, and a storage device such as a hard disk built into the computer system. Furthermore, the so-called "computer-readable recording medium" also includes a recording medium that keeps the program for a certain period of time, such as a volatile memory (RAM) inside a computer system that becomes a server or a client in the case of sending a program via a network such as the Internet or a communication line such as a telephone line.

[0132] In addition, the above-mentioned program can be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or a transmission wave in the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium having the function of transmitting information, such as a network (communication network) such as the Internet, a communication line (communication line) such as a telephone line, etc. In addition, the above-mentioned program can also be used to realize a part of the aforementioned functions. Furthermore, it can also be a program that can realize the aforementioned functions by combining with a program already recorded in a computer system, i.e., a so-called differential file (differential program).

[0133] Explanation of symbols

[0134] 100…Material Recommendation Device

[0135] 110 ... parameter information storage unit

[0136] 111…Material parameter information

[0137] 112…Emotional language parameter information

[0138] 113…Modal correction information

[0139] 120…Input

[0140] 121…Product Category Input

[0141] 122…Emotional language input unit

[0142] 123…Usage status input section

[0143] 131…Material Parameter Extraction Department

[0144] 132…Emotional language parameter extraction unit

[0145] 133…Modal correction unit

[0146] 134…Recommendation Materials Decision Department

[0147] 140…output section.

Claims

1. A material recommendation device, comprising: An input unit that receives input of information indicating a product category and emotional language, wherein the emotional language indicates a language that expresses emotions; a material parameter extraction unit for extracting a value of a sensory parameter for each material corresponding to the product category input to the input unit, based on material parameter information that establishes a correspondence between materials used in products belonging to the product category and values ​​of sensory parameters indicating the degree of sensation felt by a person with respect to the materials; an emotional language parameter extraction unit, which extracts the value of the feeling parameter corresponding to the emotional language input to the input unit based on the emotional language parameter information that establishes a correspondence between the emotional language and the value of the feeling parameter; and A recommended material determination unit determines the material recommended for the emotional language input to the input unit based on the value of the sense parameter of each of the materials extracted by the material parameter extraction unit and the value of the sense parameter extracted by the emotional language parameter extraction unit.

2. The material recommendation device according to claim 1, wherein: The emotional language is a sentence classified into an evaluative factor or an active factor among the factors of the EPA structure defined in the impression evaluation based on the semantic differential (SD method), namely, evaluation, efficacy, and activity.

3. The material recommendation device according to claim 1 or 2, wherein: The material recommendation device further includes a correction unit that corrects the value of the sense parameter using a coefficient that takes into account a cross-modal effect in which impressions respectively felt with a plurality of senses are correlated with each other.

4. The material recommendation device according to claim 3, wherein: The coefficients are set to different values ​​according to the order in which the impressions are felt through the plurality of senses. The correction unit acquires information indicating the order, and performs correction using the coefficient corresponding to the acquired order.

5. The material recommendation device according to claim 4, wherein: In the coefficient for correcting the value of the sensory parameter for each of the materials, a value of 1 is set in the case of the first sense, a value greater than 0 and less than 1 is set in the case of the later sense, and 0 is set in the case of the other senses.

6. The material recommendation device according to claim 4, wherein: In the coefficient for correcting the value of the sense parameter of the emotional language, a value of 1 is set when the sense produces the cross-modal effect, and a value of 0 is set when the sense does not produce the cross-modal effect.

7. The material recommendation device according to claim 1 or 2, wherein: The senses refer to the five senses or senses further subdivided from the five senses.

8. A material recommendation method, comprising: An input step, receiving input of information representing product categories and emotional language, where the emotional language represents language for expressing emotions; a material parameter extraction step of extracting a value of a sensory parameter of each material corresponding to the product category input in the input step, based on establishing corresponding material parameter information between materials used in products belonging to the product category and values ​​of sensory parameters indicating the degree of sensation felt by a person with respect to the materials; an emotional language parameter extraction step, extracting the value of the feeling parameter corresponding to the emotional language input in the input step based on establishing corresponding emotional language parameter information between the emotional language and the feeling parameter value; and The recommended material determination step determines the material recommended for the emotional language input in the input step based on the value of the sensory parameter of each material extracted in the material parameter extraction step and the value of the sensory parameter extracted in the emotional language parameter extraction step.

9. A program for causing a computer to execute the following steps: An input step, receiving input of information representing product categories and emotional language, where the emotional language represents language for expressing emotions; a material parameter extraction step of extracting a value of a sensory parameter of each material corresponding to the product category input in the input step, based on establishing corresponding material parameter information between materials used in products belonging to the product category and values ​​of sensory parameters indicating the degree of sensation felt by a person with respect to the materials; an emotional language parameter extraction step, extracting the value of the feeling parameter corresponding to the emotional language input in the input step based on establishing corresponding emotional language parameter information between the emotional language and the feeling parameter value; and The recommended material determination step determines the material recommended for the emotional language input in the input step based on the value of the sensory parameter of each material extracted in the material parameter extraction step and the value of the sensory parameter extracted in the emotional language parameter extraction step.

Images