Information processing system, information processing method, and computer-readable medium

Abstract

The present invention provides an information processing system, an information processing method, and a computer-readable medium that more accurately present fragrances corresponding to individual differences in preferences and reactivity of a subject with less burden. The information processing system includes a classification unit, an acquisition unit, a selection unit, an evaluation unit, and a generation unit. The classification unit classifies a plurality of fragrances into segments specified in a manner corresponding to a combination of a first classification item on a first classification axis and a second classification item on a second classification axis. The acquisition unit receives input of information associated with the first classification axis from the subject. The selection unit determines the first classification item based on the information associated with the first classification axis, and selects a fragrance for evaluation from each of the segments corresponding to the first classification item and having different second classification items. The evaluation unit acquires fragrance evaluation information from the subject for each of the selected fragrances for evaluation. The generation unit generates fragrance information for the fragrance to be presented based on the evaluation information.

Description

Technical Field

[0001] The present invention relates to an information processing system, an information processing method, and a computer-readable medium for presenting a scent corresponding to a subject. Background Art

[0002] Currently, technologies have been proposed that automatically suggest appropriate fragrances tailored to individual preferences or circumstances without requiring expert knowledge or technical expertise. For example, Japanese Patent No. 6665899 discloses a fragrance suggestion information output system that clusters users based on fragrance evaluation information input by users and then recommends fragrances based on the classifications. Summary of the Invention

[0003] However, the system described in Japanese Patent No. 6665899 cannot classify users into appropriate clusters when there are only a few types of fragrances to be evaluated, or when the fragrances to be evaluated tend to be similar. On the other hand, if a large number of fragrances are to be evaluated in order to achieve appropriate cluster classification, the evaluation process would require considerable time, placing an increased burden on the user.

[0004] The present invention has been made to solve such problems, and its object is to provide an information processing system, an information processing method, and a program that can more accurately present a scent corresponding to individual differences in preferences and responsiveness of a subject with less burden.

[0005] An information processing system according to one embodiment of the present invention presents a fragrance to a subject. The information processing system includes a classification unit, an acquisition unit, a selection unit, an evaluation unit, and a generation unit. The classification unit classifies a plurality of fragrances into segments defined to correspond to a combination of a first classification item on a first classification axis and a second classification item on a second classification axis. The acquisition unit receives input of information related to the first classification axis from the subject. The selection unit determines a first classification item based on the information related to the first classification axis and selects at least one fragrance for evaluation from each segment corresponding to the first classification item and having different second classification items. The evaluation unit acquires fragrance evaluation information from the subject for each selected fragrance for evaluation. The generation unit generates fragrance information for a fragrance to be presented based on the evaluation information. With this configuration, the subject only needs to evaluate the fragrances narrowed down based on the first classification item, thereby reducing the evaluation burden. Furthermore, the information processing system can accurately select fragrances for presentation that are tailored to the subject's preferences and individual responsiveness from a diverse set of fragrances with different second classification items, and present them to the subject.

[0006] Here, the first classification item can be related to the intended use of the spice. This allows spices to be appropriately categorized based on their intended use. Alternatively, the second classification item can be related to the flavor quality of the spice. This ensures a diverse range of spices for evaluation.

[0007] The classification unit can determine the first classification item of each of the multiple spices based on the degree of matching with each item on the first classification axis. Furthermore, the classification unit can determine, for each segment, the spices classified into the segment that have the highest degree of matching with the first classification item corresponding to the segment as the expression fragrance representing the segment. The selection unit can select, as the evaluation fragrance, the expression fragrance of each segment corresponding to the first classification item determined based on the information associated with the first classification axis and having different second classification items. In this way, the diversity of evaluation spices is more effectively ensured, so that the information processing system can more accurately select prompt spices and provide prompts to the subject.

[0008] Furthermore, the selection unit may select, as the evaluation fragrance, an aroma whose degree of matching with the first classification item determined based on information associated with the first classification axis exceeds a predetermined threshold. This can reduce the number of fragrances evaluated by the subject, further alleviating the evaluation burden.

[0009] Alternatively, the evaluation information may include an evaluation score, and the evaluation unit may prompt the subject to input evaluation scores, starting with the fragrance for evaluation that has a higher degree of match with the first classification item determined based on the information associated with the first classification axis. Furthermore, the evaluation unit may terminate acceptance of the input of the evaluation information if the evaluation score falls below a predetermined threshold, or if the evaluation score falls below the evaluation score of the fragrance for evaluation that was previously input by at least a predetermined threshold. This allows the subject to terminate the evaluation early, thereby further reducing the evaluation burden on the subject.

[0010] The evaluation information may include an evaluation score, and the evaluation unit may determine as a presentation fragrance an evaluation fragrance having an evaluation score exceeding a predetermined threshold. Furthermore, the evaluation unit may determine as a presentation fragrance a fragrance obtained by mixing each of the evaluation fragrances having an evaluation score exceeding the predetermined threshold based on the evaluation score. In this manner, the information processing system can present a more suitable fragrance to the subject.

[0011] Alternatively, the evaluation unit may infer fragrance evaluation information based on the subject's biological information. This allows evaluation processing to be performed even without the subject inputting an evaluation score, thereby reducing the evaluation burden on the subject. Furthermore, the information processing system improves evaluation accuracy by utilizing biological information, including unconscious physiological responses, thereby enabling more accurate selection of fragrances for presentation.

[0012] An information processing method according to one embodiment of the present invention is an information processing method for presenting a fragrance to a subject. The information processing method includes a classification stage, an acquisition stage, a selection stage, an evaluation stage, and a generation stage. The classification stage is a stage for classifying a plurality of fragrances into segments defined in accordance with a combination of a first classification item on a first classification axis and a second classification item on a second classification axis. The acquisition stage is a stage for accepting input of information associated with the first classification axis from the subject. The selection stage is a stage for determining a first classification item based on the information associated with the first classification axis and selecting a fragrance for evaluation from each segment corresponding to the first classification item and having a different second classification item. The evaluation stage is a stage for obtaining fragrance evaluation information from the subject for each of the selected fragrances for evaluation. The generation stage is a stage for generating fragrance information for presentation to the subject based on the evaluation information. With this structure, the subject only needs to evaluate the fragrances narrowed down based on the first classification item, thereby reducing the evaluation burden. In addition, a fragrance for presentation that matches individual differences in preferences and responsiveness of the subject can be accurately selected from fragrances whose second classification items are different and whose diversity is ensured, and presented to the subject.

[0013] A program according to one embodiment of the present invention is a program for presenting a fragrance to a subject. The program causes a computer to execute classification, acquisition, selection, evaluation, and generation processing. The classification process is a process for classifying a plurality of fragrances into segments defined in accordance with a combination of a first classification item on a first classification axis and a second classification item on a second classification axis. The acquisition process is a process for accepting input of information associated with the first classification axis from the subject. The selection process is a process for determining a first classification item based on the information associated with the first classification axis and selecting a fragrance for evaluation from each segment corresponding to the first classification item and having a different second classification item. The evaluation process is a process for acquiring fragrance evaluation information from the subject for each of the selected fragrances for evaluation. The generation process is a process for generating fragrance information for presentation to the subject based on the evaluation information. With this configuration, the subject only needs to evaluate the fragrances narrowed down based on the first classification item, thereby reducing the evaluation burden. In addition, a fragrance for presentation that matches individual differences in preferences and responsiveness of the subject can be accurately selected from fragrances whose second classification items are different and whose diversity is ensured, and presented to the subject.

[0014] According to the present invention, it is possible to provide an information processing system, an information processing method, and a program that can more accurately present a fragrance corresponding to individual differences in preferences and responsiveness of a subject with less burden.

[0015] The above and other objects, features and advantages of the present invention will become more fully understood from the detailed description given below and the accompanying drawings, which are for illustration only and thus should not be considered as limiting the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic structural diagram of the aroma presentation system according to the first embodiment.

[0017] Figure 2 This is a diagram for explaining an example of the classification table according to the first embodiment.

[0018] Figure 3 This is a block diagram showing the hardware configuration of the information processing device according to the first embodiment.

[0019] Figure 4 This is a flowchart showing the procedure of the classification process according to the first embodiment.

[0020] Figure 5 This is a diagram for explaining the purpose of use information according to the first embodiment.

[0021] Figure 6 This is a diagram for explaining the classification process of spices according to the first embodiment.

[0022] Figure 7 This is a diagram for explaining the process of determining the expression flavor according to the first embodiment.

[0023] Figure 8 This is a diagram for explaining the process of determining the expression flavor according to the first embodiment.

[0024] Figure 9 This is a diagram for explaining the process of determining the expression flavor according to the first embodiment.

[0025] Figure 10 This is a diagram for explaining the process of determining the expression flavor according to the first embodiment.

[0026] Figure 11 This is a diagram for explaining the process of determining the expression flavor according to the first embodiment.

[0027] Figure 12 This is a diagram for explaining the process of determining the expression flavor according to the first embodiment.

[0028] Figure 13 This is a diagram for explaining the process of determining the expression flavor according to the first embodiment.

[0029] Figure 14 This is a flowchart showing the procedure of information processing during operation of the information processing device according to the first embodiment.

[0030] Figure 15 This is a flowchart showing the procedure of the evaluation process according to the second embodiment.

[0031] Figure 16 This is a flowchart showing the procedure of the evaluation process and the determination process according to the third embodiment.

[0032] Figure 17 This is a schematic structural diagram of the aroma presentation system according to the fourth embodiment.

[0033] Figure 18 A diagram for explaining the prediction process according to the fourth embodiment.

[0034] Figure 19 This is a diagram showing an example of a radar chart showing the main component fractions of a fragrance according to the fourth embodiment.

[0035] Figure 20 This is a schematic structural diagram of the aroma presentation system according to the fifth embodiment. DETAILED DESCRIPTION

[0036] Hereinafter, although the present invention is described by way of embodiments, the invention to which the scope of patent protection relates is not limited to the following embodiments. In addition, all structures described in the embodiments are not necessarily necessary as means for solving the problem. For the purpose of clear explanation, the following description and drawings are appropriately omitted and simplified. In addition, in each of the drawings, the same symbols are used to mark the same elements.

[0037] <Issues of implementation>

[0038] Here, the problem to be solved by this embodiment will be described again.

[0039] In recent years, customized perfumes (fragrances) corresponding to personal preferences or situations have attracted attention. For example, the following service is currently proposed, that is, the subject smells various spices and spices mixed together to decide the fragrance formula suitable for preferences and usage scenarios, and purchases the original fragrance formula on the spot. However, in order to properly decide the fragrance formula, expert advice is required. Therefore, when the climate and purpose of using the fragrance are different, it is necessary to discuss with the expert every time. In addition, when selecting sample fragrances in order to investigate the trend of the subject's preferences, or when understanding the individual's preferences or reactivity, professional knowledge or technical know-how will be required. Therefore, there is a demand for a fragrance prompting device that automatically prompts the appropriate fragrance corresponding to the individual's preferences or situation without the need for expert knowledge or technical know-how.

[0040] For example, a method for changing the concentration or type of fragrance based on the pressure load generated by the fragrance detected according to the skin temperature has been proposed. Here, it is known that fragrance can cause emotional reactions and physiological reactions, and thus it is expected to improve the effects of comfort, relaxation, and physical and mental activation (refreshment). In addition, among the emotions evoked by fragrance, there are innate emotions and emotions formed through experience, memory, knowledge, etc., and the latter in particular vary from person to person. Therefore, it is relatively difficult to predict an individual's preferences and physiological / psychological reactivity in advance. Therefore, in such a method, even if a fragrance with a lower pressure load can be selected, it is relatively difficult to select a suitable fragrance that fully meets the individual differences in physiological and psychological reactivity to the fragrance.

[0041] Furthermore, as described in Japanese Patent No. 6665899, a method is proposed for clustering users based on their input fragrance evaluation information and recommending fragrances based on these classifications. However, if the number of fragrances to be evaluated is small, or if the fragrances to be evaluated tend to be similar, it may be difficult to categorize users into appropriate clusters. On the other hand, if a large number of fragrances are to be evaluated in order to achieve appropriate clustering, the evaluation process would require considerable time, placing an increased burden on the user. Furthermore, there are many different types of fragrance substances, and their number is practically infinite, considering their types and blending methods (e.g., mixing ratios). Currently, there is no established system for quantitatively expressing the sense of smell (i.e., olfaction) similar to the three primary colors in vision, and there is no unified standard for categorizing fragrances. Japanese Patent No. 6665899 also does not disclose how to categorize users into clusters and how to indicate which fragrance they are evaluating.

[0042] As described above, conventional scent presentation devices have problems in terms of burden on the recipient and accuracy. The present disclosure is an invention made to solve these problems, and embodiments are described below.

[0043] <Implementation Method 1>

[0044] First, use Figures 1 to 14 , implementation mode 1 is described. Figure 1 This is a schematic diagram of a scent presentation system 1 according to Embodiment 1. The scent presentation system 1 is a computer system that presents a scent corresponding to a subject P to a subject P. For example, the scent presentation system 1 is installed in a store or accommodation facility. The scent presentation system 1 includes an information processing device (information processing system) 10, a display input device 30, and a supply device 40.

[0045] The information processing device 10 is a computer device that performs information processing for presenting a scent suitable for the subject P to the subject P. The information processing device 10 is connected to a display input device 30 and a supply device 40. Upon receiving information related to a first classification axis (described later) via the display input device 30, the information processing device 10 causes the subject P to evaluate each of a plurality of evaluation fragrances via the supply device 40. Based on the evaluation results, the information processing device 10 determines a fragrance to present to the subject P and presents the fragrance to the subject P via the display input device 30 or the supply device 40.

[0046] The display input device 30 is a display input device such as a touch panel. The display input device 30 receives input from the subject P and provides the received information to the information processing device 10. The display input device 30 also displays the information provided by the information processing device 10. Alternatively, the display input device 30 may be composed of a display device such as a monitor and an input device such as a mouse and a keyboard.

[0047] The supply device 40 is a dispenser that contains multiple fragrances and produces fragrance using one or more selected fragrances from the stored fragrances. The supply device 40 sprays or sprays (dispenses) the fragrance for evaluation under control from the information processing device 10. Furthermore, the supply device 40 supplies the fragrance for presentation under control from the information processing device 10.

[0048] Next, the functional configuration of the information processing device 10 will be described. The information processing device 10 includes a classification unit 11 , an acquisition unit 12 , a selection unit 13 , an evaluation unit 14 , a generation unit 15 , and a flavor database (DB) 20 .

[0049] The classification unit 11 is connected to the fragrance DB 20 and classifies a plurality of fragrances into segments set using a plurality of classification axes based on the data stored in the fragrance DB 20. The plurality of fragrances are natural fragrances, synthetic fragrances or blended fragrances. The classification axis is an evaluation axis for classifying fragrances. In embodiment 1, the segments are set in a manner corresponding to the combination of the first classification item on the first classification axis and the second classification item on the second classification axis. The first classification item is one of the explanatory variables set along the first classification axis. Preferably, the first classification axis and the first classification item are evaluation axes and explanatory variables that the subject can objectively identify and have a high ability to classify fragrances. In embodiment 1, the first classification axis is associated with the purpose of use of the fragrance. In addition, the first classification item is an item (purpose of use item) associated with the purpose of use of the fragrance, for example, "comfort (purpose)", "relaxation (purpose)", "refreshment (purpose)" and "waking up (purpose)" can be listed. In this way, fragrances can be appropriately classified according to the purpose of use.

[0050] The second classification item is one of the explanatory variables set along the second classification axis. Preferably, the second classification axis and the second classification item are evaluation axes and explanatory variables suitable for ensuring flavor diversity. In Embodiment 1, the second classification axis is an axis associated with the flavor quality of spices. Furthermore, the second classification item is an item associated with the flavor quality of spices, and examples thereof include "citrus," "woody," "floral," "sweet," and "mint-vanilla."

[0051] The classification unit 11 classifies a plurality of fragrances into segments in advance, creates a classification table T, and stores the classification table in the fragrance DB 20 .

[0052] The acquisition unit 12 is connected to the display input device 30 and receives input of information related to the first classification axis from the subject P via the display input device 30. In the first embodiment, the information related to the first classification axis is information related to the purpose of use.

[0053] The selection unit 13 determines a first classification item for the evaluation fragrance to be used in subsequent processing based on the information associated with the first classification axis acquired by the acquisition unit 12. Furthermore, the selection unit 13 uses the classification table T to select an evaluation fragrance from each segment corresponding to the first classification item and having a different second classification item.

[0054] The evaluation unit 14 is connected to the display input device 30 and the supply device 40 and performs evaluation processing for the evaluation fragrance. Specifically, the evaluation unit 14 prompts the subject P to evaluate the selected evaluation fragrance via the display input device 30. Next, the evaluation unit 14 transmits information about the selected evaluation fragrance to the supply device 40, causing the supply device 40 to spray or atomize the evaluation fragrance. Next, the evaluation unit 14 obtains fragrance evaluation information from the subject P via the display input device 30, i.e., accepts input of fragrance evaluation information for each selected evaluation fragrance. Furthermore, the evaluation unit 14 determines a fragrance to be presented based on the evaluation information.

[0055] The generator 15 generates fragrance information for the determined reminder fragrance. The fragrance information may include identification information (ID) of the fragrance and information on the mixing ratio when the reminder fragrance is a blending fragrance. The generator 15 is connected to the supply device 40 and transmits the fragrance information for the reminder fragrance to the supply device 40, causing the supply device 40 to spray or atomize the reminder fragrance. In addition, the generator 15 is connected to the display input device 30 and causes the display input device 30 to display the fragrance information for the reminder fragrance. By adopting this method, the generator 15 can present the reminder fragrance to the subject P.

[0056] The fragrance DB 20 is a storage medium that stores information required for information processing by the information processing device 10. For example, the fragrance DB 20 stores information on a plurality of fragrances, including data used in the classification processing performed by the classification unit 11. Furthermore, the fragrance DB 20 stores the classification table T generated by the classification unit 11.

[0057] Furthermore, the information processing device 10 may further include a sales processing unit (not shown) for performing calculation processing for selling the presentation fragrance to the subject person P and management processing of sales information.

[0058] In addition, although the fragrance DB 20 is included in the information processing device 10 in the present embodiment 1, the fragrance DB 20 may be included in an external device (not shown) connected to the information processing device 10 in a communicable manner instead.

[0059] Figure 2 This figure illustrates an example of a classification table T according to Embodiment 1. In Embodiment 1, the classification table T is a two-dimensional table defined with the classification items of purpose as the first classification axis as columns and the classification items of taste as the second classification axis as rows.

[0060] In this figure, the classification items of purpose of use are "comfort", "relaxation" and "refreshment", and the segment groups located in the columns specifically specified by these classification items are called column groups C1 to C3. In addition, in this figure, the classification items of flavor are "citrus", "woody", "floral", "sweet" and "vanilla mint", and the segment groups located in the rows specified by these classification items are called row groups R1 to R5. Although the number of columns (i.e., the number of column groups) and the number of rows (i.e., the number of row groups) are not limited to this, they are at least two or more. In particular, the more row groups there are, the more diversity of fragrance can be ensured, but on the other hand, the evaluation burden on the subject will increase. Therefore, the number of row groups is set to an appropriate number by comparing the benefits generated by diversity and the risks of increased evaluation burden. In addition, the number of row groups can be set according to the diversity of fragrances registered in the fragrance DB20 and the objects of the fragrance prompts. For example, when the number of types of fragrances registered in the fragrance DB 20 is about 80 and the purpose is to present fragrances to general users, the appropriate number of row groups is about 7 (6 to 8).

[0061] For example, a fragrance classified as "comfort" in the first classification item and "citrus" in the second classification item is "grapefruit essential oil." "Grapefruit essential oil" is registered in the segment belonging to column group C1 and row group R1 in classification table T. Furthermore, a fragrance classified as "comfort" in the first classification item and "tree" in the second classification item is "frankincense essential oil." "Frankincense essential oil" is registered in the segment belonging to column group C1 and row group R2.

[0062] The fragrance registered in a segment is a fragrance called an expression fragrance that represents the segment. The expression fragrance is selected from fragrances classified into the segment as a fragrance that appropriately expresses the first and second classification items of the segment (described later).

[0063] Figure 3This is a block diagram showing the hardware configuration of the information processing device 10 according to the first embodiment.

[0064] The information processing device 10 mainly includes a processor 100, a ROM 101 (Read Only Memory), a RAM 102 (Random Access Memory), and an interface 103 (IF). The processor 100, ROM 101, RAM 102, and interface 103 are interconnected via a data bus or the like.

[0065] The processor 100 has a function as a computing device that performs control processing and calculation processing. The processor 100 can be a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an FPGA (field-programmable gate array), a DSP (digital signal processor) or an ASIC (application specific integrated circuit) or a combination thereof. The ROM 101 has a function for storing control programs and calculation programs executed by the processor 100. The RAM 102 has a function for temporarily storing processing data, etc. The interface unit 103 inputs and outputs signals with the outside via wired or wireless communication. In addition, the interface unit 103 receives data input operations performed by the user and displays information to the user. For example, the interface unit 103 communicates with the display input device 30 and the supply device 40.

[0066] Next, use Figure 4 , and refer to Figures 5 to 13 Next, the classification process performed by the information processing device 10 will be described. Figure 4 This is a flowchart showing the procedure of the classification process according to Embodiment 1. This classification process is performed before the operation of the information processing device 10. However, this classification process may also be performed when the fragrance information in the fragrance DB 20 is updated during operation.

[0067] First, in Figure 4 In step S10, the classification unit 11 of the information processing device 10 sets which fragrance is to be classified, that is, the fragrance to be classified. The classification unit 11 may set all fragrances stored in the fragrance DB 20 as the classification target, or may set a part of the fragrances as the classification target.

[0068] Next, in step S11, the classification unit 11 obtains the usage purpose information for each of the fragrances to be classified from the fragrance DB 20. The usage purpose information is pre-generated based on the results of a large number of test subjects evaluating the fragrance using usage purpose as an evaluation axis. In this first embodiment, the usage purpose information is information on the degree of matching with each first classification item.

[0069] Here, Figure 5 This is a diagram for explaining the purpose of use information involved in Implementation 1. Figure 5 , an example of an evaluation table A used by a large number of test subjects to evaluate fragrances with the purpose of use as the evaluation axis is shown. The purpose of use is closely linked to the emotional evaluation of the fragrance, and various emotional evaluation items are recorded in the evaluation table A. Emotional evaluation items include, for example, "pleasant / unpleasant", "like / dislike", "calm down", "feel good", "dizzy", "relax", "refreshing", and "nostalgic". In addition, the evaluation items "pleasant / unpleasant" and "like / dislike" are associated with the first classification item "comfort", the evaluation items "calm down" and "relax" are associated with the first classification item "relax", and the evaluation items "feel good" and "refreshing" are associated with the first classification item "refreshing". The test subject evaluates the purpose of use of the fragrance by smelling the fragrance and inputting a score indicating the degree of match between the fragrance and its evaluation item for each evaluation item. Although the degree of match is shown in seven levels in this figure, it is not limited to this. In addition, the classification unit 11 obtains the average degree of match with the evaluation items of a large number of test subjects and calculates the average value as the degree of match with each first classification item. For example, the classifier 11 may set the average of the degrees of matching with the evaluation item "pleasure / unpleasure" as the degree of matching with the first classification item "comfort." Alternatively, the classifier 11 may set the average of the average of the degrees of matching with the evaluation item "pleasure / unpleasure" and the average of the degrees of matching with the evaluation item "like / dislike" as the degree of matching with the first classification item "comfort." The information on the degree of matching with each first classification item calculated in this manner is stored in the fragrance DB 20 as usage purpose information, associated with the fragrance.

[0070] Back to Figure 4In step S12, the classification unit 11 obtains the taste information of each of the spices to be classified from the spice DB 20. The taste information is information that is pre-generated based on the results of a large number of test subjects evaluating the spices using taste as an evaluation axis. In the first embodiment, the taste information is information on the degree of matching of the spices with the second classification item. The test subject evaluates the taste of the spices by smelling the spices and inputting a score indicating the degree of matching of the spices with the evaluation item corresponding to the second classification item, similar to the case of the purpose of use. The evaluation items here are, for example, "citrus", "tree" or "sweet", which are also called taste evaluation words. Furthermore, the classification unit 11 calculates the average degree of matching with the evaluation item for a large number of test subjects, and sets the average value as the degree of matching with the second classification item. This process can be implemented before step S11 or in parallel with step S11.

[0071] In step S13 , the classifying unit 11 classifies the spices to be classified into segments set using two classification axes, ie, purpose of use and taste.

[0072] Here, Figure 6 FIG is a diagram for explaining the classification process of spices involved in the first embodiment. Figure 6 , shows a classification table of fragrances after classification processing and before the aroma expression determination processing described later. In this figure, each segment shows the identification number (ID) of the classified fragrance and the degree of matching M1, M2, and M3 of the fragrance with the first classification item corresponding to the segment.

[0073] First, the classification unit 11 determines the second classification item of each of the spices in the classification object based on the degree of matching with the second classification item, and classifies the spices of the classification object into row groups R1 to R5. In this figure, spices 110 to 112, spices 210 to 212, and spices 310 to 312 are classified into row group R1. In addition, the classification unit 11 determines the first classification item of each of the spices in the classification object based on the degree of matching with each item on the first classification axis, and classifies the spices of the classification object into column groups C1 to C3. In this figure, spices 110 to 112, spices 120 to 122, ... spices 150 to 152 are classified into column group C1. Thus, one or more spices will be classified into each segment. In addition, in the classification process into the above-mentioned row groups or column groups, well-known clustering methods such as the k-means clustering algorithm can be used.

[0074] Back to Figure 4In step S14, the classification unit 11 determines, for each segment, an expression fragrance from the fragrances classified into the segment based on the degree of matching with the first classification item. Here, it is preferred that the expression fragrance is a fragrance with high robustness, such as being stably classified into the same segment even if the number of segments, the number of subjects in steps S11 to 12, or the attributes of the subjects change more or less. However, the like or dislike of the fragrance, the reactivity to the fragrance, and the labeling method of the degree of matching vary from one subject to another, so how to determine a fragrance with high robustness becomes a problem. In this embodiment 1, in order to appropriately set the preferred expression fragrance, the classification unit 11 determines, for each segment, the fragrance that has the highest degree of matching with the first classification item corresponding to the segment among the fragrances classified into the segment. For example, in Figure 6 In the table, the fragrance expressed by the segment C1×R1 is the fragrance 110 (M1=9) with the highest degree of matching M1 with "comfort." The fragrance expressed by the segment C1×R2 is the fragrance 121 (M1=8) with the highest degree of matching M1 with "comfort."

[0075] Although the degree of matching with the first classification item is the average value of the degree of matching with the evaluation item corresponding to the purpose of use of the first classification item, the reason for preferably expressing the fragrance as the fragrance with the highest average value of the degree of matching with the evaluation item (highest average value fragrance) is explained in this article.

[0076] Figures 7 to 13 This is a diagram for explaining the process of determining the expression flavor according to the first embodiment. Figures 7 and 8 、 Figures 9 and 10 as well as Figures 11 and 12 These respectively correspond to the evaluation items of emotions: "pleasure / unpleasantness", "relaxation", and "refreshment". Figure 7 、 Figure 9 as well as Figure 11 It represents the relationship between the degree of matching evaluated by each subject for the fragrance with the highest average value of the degree of matching with the evaluation items of the corresponding emotions (the fragrance with the highest average value), and the degree of matching evaluated by the subject with the highest score (the fragrance with the highest score) among all the fragrances classified into the same segment. Figure 8 、 Figure 10 as well as Figure 12 It represents the relationship between the degree of matching evaluated by each subject for the fragrance that is not the fragrance with the highest average value (virtual fragrance with the highest average value) among all the fragrances classified into the same segment and the degree of matching with the fragrance with the highest score.

[0077] like Figures 7 to 12As shown in , both the highest average flavor and the virtual highest average flavor have a positive correlation in their matching degree with the highest scoring flavor. Figure 7 、 Figure 9 as well as Figure 11 As shown in the figure, the correlation coefficients r between the matching degree of the highest average flavor and the matching degree of the highest scoring flavor for "pleasant / unpleasant", "relaxing", and "refreshing" are 0.77, 0.77, and 0.80, respectively, which can be said to be a strong positive correlation. Figure 8 、 Figure 10 as well as Figure 12 As shown, the correlation coefficients r between the matching degree of the virtual highest average spices and the matching degree of the highest scoring spices for "pleasant / unpleasant", "relaxation" and "refreshing" are 0.50, 0.58 and 0.50 respectively. Compared with the highest average spices, the strength of the correlation is weaker.

[0078] Although Figure 13 In the figure, the statistical values between multiple evaluation items are shown, which are the correlation coefficients of the highest average flavor and the virtual highest average flavor. However, it can be seen from the figure that the highest average flavor and the highest scoring flavor show a stronger correlation than the virtual highest average flavor.

[0079] In this way, by having the classification unit 11 decide on the fragrance with the highest average value, that is, the fragrance with the highest degree of match with the first classification item corresponding to the segment, the diversity of fragrances used for evaluation is more effectively guaranteed, thereby improving the accuracy of selecting fragrances for prompts in subsequent applications.

[0080] Back to Figure 4 In step S15, the classifier 11 generates a classification table T based on the determined aroma expression. Furthermore, in addition to registering aroma expression information for each segment, the classifier 11 may also register information indicating the degree of matching with the first classification item corresponding to the segment. The classifier 11 stores the generated classification table T in the fragrance DB 20 and terminates the classification process.

[0081] Figure 14 This is a flowchart showing the procedure of information processing during operation of the information processing device 10 according to the first embodiment. Figure 14The steps shown include obtaining information associated with the first classification axis (step S1), selecting a fragrance for evaluation (step S2), evaluating (step S3), determining a fragrance for presentation (step S4), and generating and presenting fragrance information (step S5). In this first embodiment, step S1 consists of step S20, step S2 consists of steps S21-22, step S3 consists of steps S23-25, step S4 consists of step S26, and step S5 consists of steps S27-28. The specific processes of steps S20-28 are described below.

[0082] First, in step S20, the subject P inputs information related to the purpose of using the fragrance. For example, the subject P can select their purpose from "comfort," "relaxation," and "refreshment." The acquisition unit 12 of the information processing device 10 then acquires information related to the subject P's purpose of use. The acquisition unit 12 supplies the acquired information to the selection unit 13.

[0083] Next, in step S21, the selection unit 13 determines a first classification item based on information related to the intended use, and then determines a column group in the classification table T. Then, in step S22, the selection unit 13 uses the classification table T to select the expressed aroma for each segment in the column group as a fragrance for evaluation. Specifically, the selection unit 13 selects the expressed aroma for each segment that corresponds to the determined first classification item and has different second classification items as a fragrance for evaluation. For example, if the column group is C1, the selection unit 13 selects a total of five expressed aromas that are included in C1 and belong to segments R1 to R5 as fragrances for evaluation. This minimizes the number of fragrances evaluated by the subject P in the subsequent evaluation process while ensuring a diverse range of fragrances for evaluation, enabling accurate understanding of the user's preferences and physiological and psychological reactivity. Alternatively, the selection unit 13 may select as fragrances for evaluation those whose degree of match with the determined first classification item exceeds a predetermined threshold. This reduces the number of fragrances evaluated by the subject P and further reduces the evaluation burden. The selection unit 13 supplies information on the selected fragrance for evaluation to the evaluation unit 14 .

[0084] Next, the processes shown in steps S23 to S25 are repeatedly performed for each fragrance for evaluation.

[0085] In step S23, the evaluation unit 14 supplies the i-th (natural number) evaluation fragrance via the supply device 40. Specifically, the evaluation fragrance is adjusted in the supply device 40 so that its concentration in the air at the nose of the subject P reaches the level of "odor intensity 3 easily perceptible odor" according to the six-level odor intensity display method, and then sprayed or atomized. An example of such adjustment method is to mix and dilute air containing fragrance components with odorless (fragrance-free) air at a predetermined ratio. The subject P then smells the supplied evaluation fragrance.

[0086] In step S24, the subject P inputs an evaluation score for the i-th fragrance for evaluation that has been supplied. The evaluation score is an example of evaluation information indicating the result of evaluation by the subject P. Thus, the evaluation unit 14 of the information processing device 10 obtains the evaluation score of the fragrance for evaluation. In addition, the evaluation by the subject P can be performed using the same Figure 5 The evaluation unit 14 is implemented by using the same evaluation table A as shown in FIG. Therefore, the evaluation score can be an indicator of the same degree of matching as the evaluation item corresponding to the first classification item used in the classification process. However, the evaluation score is not limited to this. Then, the evaluation unit 14 increments the value of i. In step S25, the evaluation unit 14 determines whether i exceeds the total number n of evaluation spices, that is, whether all evaluation spices have been evaluated. If all evaluation spices have not been evaluated (no in step S25), the evaluation unit 14 returns the process to step S23. If all evaluation spices have been evaluated (yes in step S25), the process proceeds to step S26.

[0087] In step S26, the evaluation unit 14 determines a presentation fragrance based on the evaluation results. In the first embodiment, the evaluation unit 14 determines as a presentation fragrance a fragrance with an evaluation score that is greater than a predetermined threshold. However, instead of this, the evaluation unit 14 may determine as a presentation fragrance the fragrance with the highest evaluation score or fragrances ranked in the top three (e.g., the top three) of the evaluation scores. The evaluation unit 14 supplies the presentation fragrance information to the generation unit 15.

[0088] In step S27, the generator 15 generates fragrance information for the fragrance to be suggested. Then, in step S28, the generator 15 causes the supply device 40 to supply the fragrance to be suggested. Furthermore, the generator 15 causes the display input device 30 to display the fragrance information for the fragrance to be suggested. Furthermore, the generator 15 may cause the display input device 30 to display the results of an analysis of the preferences and physiological / psychological reactivity of the subject P based on the evaluation score. This makes it easier for the subject P to understand that the suggested fragrance is suitable for the subject P and to have the impression that it is a special fragrance that belongs only to them. The generator 15 then terminates the process.

[0089] Furthermore, while a single classification table T is used in the first embodiment, different classification tables T may be generated and used based on attributes such as gender, age group, or occupational classification, or usage conditions such as season, time period, or location. In this case, in step S20, the acquisition unit 12 may receive input of attribute information or usage condition information from the subject P via the display input device 30 to acquire this information. Furthermore, the selection unit 13 may select the classification table T to be used based on the attribute information or usage condition information prior to step S21.

[0090] As described above, according to Embodiment 1, the information processing device 10 narrows down the spices based on the objective input information of the subject P for the first classification item, and narrows down the spices based on the sensibility of the subject P for the second classification item. Since the subject P only needs to actually evaluate the spices narrowed down based on the first classification item, the number of evaluations is reduced, thereby reducing the evaluation burden. Since the spices narrowed down based on the first classification item are spices that are different from each other in the second classification item, diversity is ensured. Therefore, the information processing device 10 can accurately select a suggestion spice that corresponds to the individual differences in the preferences and reactivity of the subject P from among a variety of spices, and present it to the subject P.

[0091] Since the evaluation fragrances actually evaluated by the subject P are the expression aromas representing each segment, the information processing device 10 can more reliably ensure the diversity of the evaluation fragrances and more accurately select the presentation fragrances to present to the subject P.

[0092] <Implementation Method 2>

[0093] Next, use Figure 15 , Embodiment 2 will be described. The information processing device 10 according to Embodiment 2 is basically the same as the information processing device 10 according to Embodiment 1. However, in Embodiment 2, the evaluation process ( Figure 14The processing order of step S3) is different from that of implementation mode 1.

[0094] Figure 15 This is a flowchart showing the procedure of the evaluation process according to the second embodiment.

[0095] First, in step S30, the evaluation unit 14 evaluates the Figure 14 The evaluation order of the evaluation fragrances to be performed by the subject P is determined based on the degree of matching of the first classification item determined in step S2). Specifically, the evaluation unit 14 causes the subject P to evaluate the evaluation fragrances in order from the fragrances with the highest degree of matching with the first classification item. For example, when the first classification item is "comfort", the evaluation unit 14 causes the subject P to evaluate in the order of the degree of matching M1 in the column group C1 from high to low, that is, in the order of R1→R2→R4→R3→R5 (refer to Figure 6 ) Next, the processing shown in steps S31 to S34 is repeatedly performed based on the evaluation order.

[0096] In step S31, similarly to step S23, the evaluation unit 14 supplies the i-th (natural number) evaluation fragrance via the supply device 40. The subject P smells the supplied evaluation fragrance. In step S32, similarly to step S24, the subject P inputs an evaluation score for the i-th evaluation fragrance supplied. Next, in step S33, the evaluation unit 14 determines whether to end (suspend) the evaluation of the evaluation fragrance performed by the subject P. For example, the evaluation unit 14 determines to suspend the evaluation when the evaluation score is less than a predetermined threshold value, or when the evaluation score is lower than the evaluation score of the evaluation fragrance just input by more than a predetermined threshold value. When the evaluation is not suspended (No in step S33), the evaluation unit 14 increments the value of i and determines whether all the evaluation fragrances have been evaluated (step S34). If all the fragrances for evaluation have not yet been evaluated (No in step S34), the evaluation unit 14 returns the process to step S31. If all the fragrances for evaluation have been evaluated (Yes in step S34), the evaluation unit 14 ends the process. On the other hand, if the evaluation is to be stopped (Yes in step S33), the evaluation unit 14 immediately ends the process.

[0097] Furthermore, in the subsequent determination process (step S4), the evaluation unit 14 determines a presentation fragrance based on the evaluation result of the evaluated evaluation fragrance.

[0098] In this manner, the evaluation unit 14 of the information processing device 10 according to Embodiment 2 causes the subject P to input evaluation scores in order, starting with the fragrance that has the highest degree of match with the first classification item determined based on the information associated with the first classification axis. Furthermore, if the input evaluation score is low, or significantly lower than the evaluation score of the fragrance input immediately before, the evaluation unit 14 stops accepting the input of evaluation scores. This allows the evaluation performed by the subject P to be terminated early, thereby further reducing the evaluation burden on the subject P.

[0099] Alternatively, at the time that the evaluation of several fragrances by the subject P has concluded, the evaluation unit 14 may analyze the subject P's preferences or physiological or psychological reactivity based on the degree of compatibility with the first classification item and the subject P's evaluation score, and thereby change the order of the fragrances to be evaluated. For example, if the initial evaluation order is R1 → R2 → R4 → R3 → R5, and the fragrance belonging to R4 has been evaluated, the evaluation unit 14 may change the evaluation order of the remaining fragrances to R5 → R3. This allows the evaluation by the subject P to be terminated earlier, thereby further reducing the evaluation burden on the subject P.

[0100] Furthermore, although the evaluation unit 14 determines the evaluation order based on the degree of matching with the determined first classification item, the evaluation order may be determined based on the attribute information or usage condition information of the subject P in addition to or instead of this.

[0101] <Implementation Method 3>

[0102] Next, use Figure 16 , Embodiment 3 will be described. The information processing device 10 according to Embodiment 3 is basically the same as the information processing device 10 according to Embodiment 1. However, in Embodiment 3, the evaluation process and the decision process ( Figure 14 The processing order of steps S3 and S4 is similar to that of implementation mode 1.

[0103] Figure 16 This is a flowchart showing the procedure of the evaluation process and the determination process according to the third embodiment.

[0104] First, after the processing of steps S23 to S25, in step S40, the evaluation unit 14 selects a candidate for prompting fragrance based on the evaluation results. The method for selecting the candidate for prompting fragrance may be the same as the method for selecting the prompting fragrance in step S26. For example, the candidate for prompting fragrance is an evaluation fragrance having an evaluation score above a predetermined threshold value among the evaluation fragrances. The candidate for prompting fragrance may be 2 to 5 types. Then, in step S41, the evaluation unit 14 determines whether it is necessary to blend the candidate for prompting fragrance to generate the prompting fragrance. The case where the candidate for prompting fragrance needs to be blended to generate the prompting fragrance includes a case where there are multiple candidate for prompting fragrances, a case where the subject P wants a fragrance with a higher evaluation, or a case where the subject P wants to change the fragrance, etc. If blending is required (yes in step S41), the evaluation unit 14 advances the processing to step S42, and if blending is not required (no in step S41), the processing advances to step S46.

[0105] In step S42, the evaluation unit 14 determines the mixing ratio of the candidate fragrances for suggestion. The evaluation unit 14 calculates the mixing ratio by making the ratio of the evaluation scores correspond to the ratio of the sensory intensity. Here, according to the Weber / Fechner law, when the sensory intensity is set to Y, the stimulation amount is set to X, and the constants inherent in the stimulation are set to a and b, the sensory intensity is expressed by a logarithmic function such as Y=alogX+b. For example, when the first classification item is "relaxation" and the evaluation scores of fragrances A, B, and C are 4 points, 3 points, and 2 points respectively, their mixing ratio becomes e 4 :e 3 :e 2 ≈7:3:1. Therefore, the evaluation unit 14 calculates the mixing ratio so as to obtain a ratio of powers having e as the base and the fraction as the exponent.

[0106] In step S43 , the evaluation unit 14 transmits information on the mixed fragrance and the mixing ratio to the supply device 40 , and causes the supply device 40 to blend the fragrance based on the information and supply the blended fragrance to the subject P.

[0107] In step S44 , the subject P inputs an evaluation score for the blended flavor. Thus, the evaluation unit 14 of the information processing device 10 acquires the evaluation score of the blended flavor.

[0108] Next, in step S45, the evaluation unit 14 determines whether re-blending is necessary. For example, the evaluation unit 14 may determine that re-blending is necessary when the evaluation score of the blended spices is lower than the highest evaluation score of the candidate spices for prompting. In addition, even in such a case, the evaluation unit 14 may determine that re-blending is not necessary when the number of evaluations performed by the subject P exceeds a predetermined threshold. Then, when re-blending is necessary (yes in step S45), the evaluation unit 14 returns the process to step S42 and changes the mixing ratio based on the evaluation score of the blended spices and the evaluation score of the candidate spices for prompting.

[0109] On the other hand, when re-adjustment is not necessary (No in step S45 ), the evaluation unit 14 advances the process to step S46 .

[0110] In step S46, the evaluation unit 14 determines whether to use the blended fragrance or the candidate fragrance for presentation as the presentation fragrance. The evaluation unit 14 may determine whether to use either fragrance as the presentation fragrance based on the evaluation score of the blended fragrance or the evaluation score of the candidate fragrance for presentation. The evaluation unit 14 then terminates the process.

[0111] As described above, according to Embodiment 3, the evaluation unit 14 of the information processing device 10 determines a fragrance as a presentation fragrance by mixing candidate presentation fragrances according to their evaluation scores.

[0112] Furthermore, in step S43 , the evaluation unit 14 may cause the display input device 30 to display information about the prepared fragrance when causing the supply device 40 to supply the prepared fragrance to the subject P. The evaluation unit 14 may then cause the display input device 30 to display the results of an analysis of the subject P's preferences and physiological / psychological responsiveness based on the evaluation results of the subject P.

[0113] For example, the evaluation unit 14 may cause the display input device 30 to display the main fragrance based on the analysis result, and let the subject P know what kind of smell is likely to help the subject P achieve the desired state.

[0114] Furthermore, for example, based on the analysis results, the evaluation unit 14 causes the display input device 30 to display information about recommended fragrances that are best added to the main fragrance, and informs the subject P of this information. For example, if the purpose of use is "relaxation," the display input device 30 may display the suggestion, "It seems that the 'scent of wood (wood fragrance)' can help you 'relax.' If you try adding the scent of 'lavender' on a 'relaxing vacation,' it will 'feel like a retreat.'" Furthermore, the "" portion may be changed based on the evaluation results. The display input device 30 may further display, "This time's recommendation: frankincense + lavender." Furthermore, for example, if the purpose of use is "refreshment," the display input device 30 may display the suggestion, "It seems that the 'scent of citrus' can help you 'refresh.' If you try adding the scent of 'mint' on the 'morning of an important meeting,' it will 'perk you up.'" The display input device 30 may also display, "This time's recommendation: grapefruit + mint."

[0115] As a result, the subject P can easily understand that the blended fragrance to be supplied later is more appropriate, and can also easily have the impression that the blended fragrance is a special fragrance that belongs only to him or her.

[0116] <Implementation Method 4>

[0117] Next, use Figures 17-19 Embodiment 4 will now be described. In Embodiment 1, the purpose of use information used in the classification process is generated based on the results of a large number of test subjects evaluating spices using purpose of use as the evaluation axis. However, in Embodiment 4, it is characterized in that the purpose of use information used in the classification process is generated based on the results of a large number of test subjects evaluating spices using flavor as the evaluation axis.

[0118] Figure 17 This is a schematic diagram of the structure of a fragrance presentation system 1a according to Embodiment 4. The fragrance presentation system 1a according to Embodiment 4 has basically the same structure and functions as the fragrance presentation system 1 according to Embodiment 1, but differs from the fragrance presentation system 1 in that it includes an information processing device 10a in place of the information processing device 10.

[0119] The information processing device 10 a basically has the same configuration and function as the information processing device 10 , but includes a classifying unit 11 a instead of the classifying unit 11 .

[0120] The classification unit 11 a includes a prediction unit 16 in addition to the functions of the classification unit 11 .

[0121] The prediction unit 16 predicts the degree of match with the first classification item based on the degree of match with multiple categories of taste evaluation words. These multiple categories of taste evaluation words are evaluation words associated with taste and may include at least a second classification item (e.g., "citrus," "woody," "sweet"). In this fourth embodiment, there are 39 taste evaluation words. In addition to the second classification items, other items include "fishy," "pungent," and "elegant." The classifier 11a then performs classification using the degree of match with the first classification item predicted by the prediction unit 16.

[0122] Figure 18 A diagram for explaining the prediction process according to the fourth embodiment.

[0123] First, a large number of test subjects evaluate each of the classified spices and input their degree of match with the taste evaluation words, as shown in the figure. This information on the degree of match with the taste evaluation words is stored in the spice DB 20. The prediction unit 16 retrieves information on the degree of match with the taste evaluation words for each of the classified spices from the spice DB 20 (step S50).

[0124] The prediction unit 16 then converts the degree of matching with the taste evaluation words into principal component scores for each principal component axis having a smaller dimension (step S51). The principal component scores are represented by linear combinations of the degree of matching with each taste evaluation word and are calculated using the degree of matching with each taste evaluation word.

[0125] The principal component axes are calculated before the prediction process and are calculated in advance by performing principal component analysis on the degree of matching with taste evaluation words previously evaluated by subjects for various flavors (for example, 93 types).

[0126] In this fourth embodiment, the principal component axes are divided into the first through seventh principal component axes (PC1 through PC7) based on contribution rates. As an example, the first through seventh principal component axes are safety (PC1), comfort (PC2), sweetness / sharpness (PC3), citrus / fruit / tree / green (PC4), artificial / natural (PC5), rich / burnt (PC6), and floral / soap (PC7). By converting the degree of matching of taste evaluation terms into principal component scores for the principal component axes in this way, the dimensionality can be significantly reduced while maintaining the characteristics.

[0127] Then, the prediction unit 16 calculates the average value of the principal component scores of each principal component axis among the subjects for each fragrance (step S52). Specifically, the prediction unit 16 generates the following for each fragrance: Figure 19A radar chart of the mean values of the principal component scores is shown. Figure 19 An example of a radar chart showing the main component scores of a flavor. The characteristics of each flavor are represented in a heptagonal radar chart based on the average of the main component scores.

[0128] Next, the prediction unit 16 uses a prediction formula to predict the degree of matching with each first classification item based on the principal component score of the principal component axis (step S53). The prediction formula is also generated before the prediction process. The prediction formula is generated using the degree of matching with the taste evaluation words that the test subjects have previously evaluated for various spices (for example, 93 types) and the degree of matching with the first classification items that the test subjects have previously evaluated for the same spices. For example, the prediction formula is derived by performing a multiple regression analysis by a stepwise method with the degree of matching with the taste evaluation words as an explanatory variable and the degree of matching with the first classification item as the target variable.

[0129] The prediction formula for the degree of matching with the first classification items “comfort,” “relaxation,” and “refreshment” is represented as follows, for example.

[0130] [Compatibility with comfort]

[0131] =0.13+0.83×[PC2]+0.32×[PC1]-0.17×[PC6]+0.13×[PC3]+0.07×[PC5]+0.06×[PC7]

[0132] [Compatibility with relaxation]

[0133] =0.60+0.57×[PC2]-0.18×[PC4]+0.07×[PC1]+0.06×[PC6]+0.05×[PC7]+0.03×[PC3]

[0134] [Compatibility with refreshing]

[0135] =0.68+0.72×[PC2]-0.24×[PC3]+0.09×[PC1]+0.07×[PC4]-0.07×[PC5]

[0136] Here, [PC1], [PC2], ..., [PC7] are the principal component scores of PC1 to PC7, respectively. The coefficients shown before the principal component scores are standard partial regression coefficients.

[0137] In addition, different prediction formulas may be generated according to user attributes or usage conditions, and the prediction formulas may be used to predict the degree of matching with each first classification item.

[0138] Then, the classifier 11a uses the information of the predicted degree of matching with each first classification item as the purpose of use information and performs classification processing on the column groups C1 to C3. In addition, in the first embodiment, the classifier 11 uses the information of the degree of matching with each second classification item corresponding to the flavor to perform classification processing on the row groups R1 to R5. However, it is known that spices with similar flavors have the above-mentioned Figure 19 The radar chart shown in FIG. 4 is similar in shape. In the embodiment 4, the classification unit 11a can be based on the above Figure 19 The classification processing of the row groups R1 to R5 is implemented based on the shape of the radar chart.

[0139] In this way, the prediction unit 16 predicts the purpose of use information using the results of the test subject's evaluation of the flavor using the taste as the evaluation axis, thereby saving the test subject the labor and time of evaluating the flavor using the purpose of use as the evaluation axis.

[0140] <Implementation Method 5>

[0141] Next, use Figure 20 Embodiment 5 will now be described. Embodiment 5 is characterized in that, instead of the information processing device accepting input of evaluation information (evaluation score) from the subject P during evaluation processing, the information processing device acquires evaluation information based on biometric information.

[0142] Figure 20 This is a schematic diagram of the structure of a scent presentation system 1b according to Embodiment 5. The scent presentation system 1b according to Embodiment 5 has basically the same structure and functions as the scent presentation system 1a according to Embodiment 4, but differs from the scent presentation system 1a in that it includes a biological measuring device 50 and an information processing device 10b instead of the information processing device 10a.

[0143] The biological measuring device 50 measures and monitors the biological information of the subject P. This biological information is information related to the pulse or heart rate, such as heart rate and heart rate intervals. For example, the biological measuring device 50 is a pulse sensor or a heart rate sensor. Although shown in this figure as a contact type, a non-contact type is also possible. Based on the biological information of the subject P measured, the biological measuring device 50 transmits the biological information to the information processing device 10b.

[0144] The information processing device 10 b has basically the same configuration and function as the information processing device 10 a , but includes an evaluation unit 14 b in place of the evaluation unit 14 .

[0145] While the evaluation unit 14b has essentially the same structure and functions as the evaluation unit 14, it is connected to the biometrics meter 50 and estimates the fragrance evaluation score based on the received biometric information of the subject P. For example, the higher the subject P's heart rate, the lower the evaluation score estimated by the evaluation unit 14b. Furthermore, the shorter the heart rate intervals of the subject P, the lower the evaluation score estimated. Thus, the evaluation unit 14b obtains the fragrance evaluation score based on the subject P's biometric information and uses the obtained score in subsequent evaluation processing.

[0146] Alternatively, the biometric information may be information related to line of sight. In this case, the biometric measurement device 50 may be a camera that captures the face of the subject P and transmits image data of the subject P's face to the evaluation unit 14b. The evaluation unit 14b then performs image processing based on the image data received from the biometric measurement device 50 to obtain information related to line of sight. The evaluation unit 14b may then infer a low evaluation score if the amount or frequency of line of sight movement is large.

[0147] In this manner, according to Embodiment 5, the information processing device 10b performs evaluation processing based on automatically measured biometric information even when the subject P does not input an evaluation score. This reduces the evaluation burden on the subject P. Because biometric information includes the subject P's unconscious physiological reactions, the information processing device 10b can improve evaluation accuracy by utilizing biometric information and more accurately select a fragrance for presentation.

[0148] Furthermore, the information processing device 10b may also use a combination of biometric information and the evaluation score input by the subject P. For example, the evaluation unit 14b may use the evaluation score inferred based on the biometric information as the unconscious emotion evaluation score and the evaluation score input by the subject P as the conscious evaluation score for use in the evaluation process. Alternatively, the evaluation unit 14b may modify the evaluation score input by the subject P based on the biometric information.

[0149] The present invention is not limited to the above-described embodiment, and can be appropriately modified without departing from the spirit and scope of the invention.

[0150] For example, in Embodiments 1 to 5, the first classification axis is associated with the purpose of use of the fragrance, and the second classification axis is associated with the flavor of the fragrance. However, this is not limiting, and each classification axis may also be associated with the attributes of the subject or usage conditions.

[0151] In addition, in Embodiments 1 to 5, in order to minimize the evaluation burden on the subject, one expression fragrance is registered in one segment. However, this is not limited to this, and the number of types of expression fragrances registered in one segment can be set to an appropriate number according to conditions. The method for setting the number of types of expression fragrances in one segment can be the same as the method for setting the number of row groups. In this case, in step S14, the classification unit 11 determines, for each segment, the top few of the fragrances classified into the segment that match the first classification item corresponding to the segment as the expression fragrance.

[0152] Although the present invention has been described as a hardware configuration in the above embodiments, the present invention is not limited thereto and can also execute various processes related to the information processing method by causing a processor to execute a computer program.

[0153] In the above examples, the program includes a command group (or software code) for causing the computer to execute one or more functions described in the embodiment when the program is read by the computer. The program can also be stored in a non-temporary computer-readable medium or a tangible storage medium. As non-limiting examples, the computer-readable medium or tangible storage medium includes a random-access memory (RAM), a read-only memory (ROM), a cache memory, a solid-state drive (SSD) or other memory technology, a read-only compact disc (CD-ROM), a digital versatile disc (DVD), a Blu-ray (registered trademark) optical disc or other optical disc storage, a cassette tape, a magnetic tape, a magnetic disk storage or other magnetic storage device. The program can also be sent on a temporary computer-readable medium or a communication medium. As non-limiting examples, a temporary computer-readable medium or a communication medium includes an electrical, optical, acoustic or other form of transmission signal.

[0154] In the above-described embodiment, the computer is constituted by a computer system including a personal computer or a work processor. However, this is not limiting, and the computer may also be constituted by a server on a LAN (local area network), a host computer (personal computer) communicating with a host computer, or a computer system connected to the Internet. Furthermore, it is also possible to distribute functions among various devices on a network, with the entire network constituting the computer.

[0155] From the disclosure described in this manner, it will be apparent that the embodiments of the present disclosure may be varied in various ways. Such variation should not be considered as departing from the spirit and scope of the present disclosure, and it will be apparent to those skilled in the art that all such modifications are included within the scope of the appended claims.

Claims

1. An information processing system for presenting a scent to a subject, comprising: a classification unit that classifies the plurality of spices into segments specified in a manner corresponding to a combination of a first classification item on a first classification axis and a second classification item on a second classification axis; an acquisition unit that accepts input of information associated with a first classification axis from the subject; a selection unit that determines a first classification item based on information associated with the first classification axis, and selects at least one evaluation fragrance from each of the segments corresponding to the first classification item and having different second classification items; an evaluation unit that acquires fragrance evaluation information from the subject for each of the selected fragrances for evaluation; a generating unit for generating fragrance information of a fragrance for presentation based on the evaluation information, the fragrance information including identification information of the fragrance and information on a mixing ratio when the fragrance for presentation is a blending agent fragrance; The classification unit performs the following processing, namely: For each of the plurality of spices, determining a first classification item for the spice based on a degree of matching with each item on the first classification axis; For each segment, the fragrance classified into the segment that has the highest degree of matching with the first classification item corresponding to the segment is determined as the expression fragrance representing the segment. The selection unit selects, as the evaluation fragrance, an expressed aroma corresponding to each of the segments whose second classification items are different from each other and whose first classification items are determined based on information associated with the first classification axis.

2. The information processing system according to claim 1, wherein: The first classification items are items related to the purpose of use of the fragrance.

3. The information processing system according to claim 1 or 2, wherein: The second classification items are items related to the taste of spices.

4. The information processing system according to claim 1, wherein: The selection unit selects, as the evaluation fragrance, an expression aroma having a matching degree with the first classification item determined based on information associated with the first classification axis that is equal to or greater than a predetermined threshold value.

5. The information processing system according to claim 1 or 2, wherein: The evaluation information includes an evaluation score. The evaluation unit causes the subject to input evaluation scores in order from the evaluation fragrance having a higher degree of matching with the first classification item determined based on information associated with the first classification axis.

6. The information processing system according to claim 5, wherein: The evaluation unit stops accepting input of the evaluation information when the evaluation score is less than a predetermined threshold value or when the evaluation score is lower than the evaluation score of the immediately preceding evaluation fragrance by at least a predetermined threshold value.

7. The information processing system according to claim 1 or 2, wherein: The evaluation information includes an evaluation score. The evaluation unit determines, as a presentation fragrance, an evaluation fragrance having the evaluation score equal to or greater than a predetermined threshold value.

8. The information processing system according to claim 1 or 2, wherein: The evaluation information includes an evaluation score. The evaluation unit determines, as a presentation fragrance, a fragrance obtained by mixing each of the evaluation fragrances having the evaluation score equal to or greater than a predetermined threshold value based on the evaluation score.

9. The information processing system according to claim 1 or 2, wherein: The evaluation unit estimates evaluation information of the fragrance based on biological information of the subject.

10. An information processing method executed by a computer for presenting a scent to a subject, comprising: a classification stage of classifying the plurality of spices into segments specified in a manner corresponding to a combination of a first classification item on a first classification axis and a second classification item on a second classification axis; In the acquisition stage, input of information related to the first classification axis is received from the subject; In a selection stage, a first classification item is determined based on information associated with the first classification axis, and at least one fragrance for evaluation is selected from each of the segments corresponding to the first classification item and having different second classification items; In the evaluation stage, fragrance evaluation information is obtained from the subject for each of the selected fragrances for evaluation; In the generation stage, based on the evaluation information, fragrance information of a fragrance for presentation to the subject is generated, the fragrance information including identification information of the fragrance and information on a mixing ratio when the fragrance for presentation is a blending fragrance. The classification stage includes the following processing performed by the computer, namely: For each of the plurality of spices, determining a first classification item for the spice based on a degree of matching with each item on the first classification axis; For each segment, the fragrance classified into the segment that has the highest degree of matching with the first classification item corresponding to the segment is determined as the expression fragrance representing the segment. In the selection phase, the computer selects, as the fragrance for evaluation, an expressed aroma corresponding to each of the segments having different second classification items and corresponding to the first classification item determined based on information associated with the first classification axis.

11. A computer-readable medium storing a program for presenting a scent to a subject, wherein: The program causes the computer to execute the following processing, namely: a classification process of classifying the plurality of spices into segments specified in a manner corresponding to a combination of a first classification item on a first classification axis and a second classification item on a second classification axis; an acquisition process of accepting input of information associated with the first classification axis from the subject; a selection process of determining a first classification item based on information associated with the first classification axis, and selecting at least one evaluation fragrance from each of the segments corresponding to the first classification item and having different second classification items; an evaluation process of acquiring fragrance evaluation information from the subject for each of the selected fragrances for evaluation; generating, based on the evaluation information, fragrance information of a fragrance for presentation to the subject, the fragrance information including identification information of the fragrance and information on a mixing ratio when the fragrance for presentation is a blending fragrance; The classification process includes the following steps, namely: For each of the plurality of spices, determining a first classification item for the spice based on a degree of matching with each item on the first classification axis; For each segment, a step of determining, among the fragrances classified into the segment, the fragrance that has the highest degree of matching with the first classification item corresponding to the segment as the expression aroma representing the segment, The selection process includes the step of selecting, as the fragrance for evaluation, an expression aroma corresponding to each of the segments having different second classification items and corresponding to the first classification item determined based on information associated with the first classification axis.

Images