Wine recommendation system and wine recommendation method

Abstract

To enable wine suited to a user's preference to be recommended even if the user does not have specialized knowledge about wine.SOLUTION: A system according to one aspect of the present disclosure includes: an acquisition unit that acquires fruit preference information relating to a user's preference for fruit-based food and drink, and fruit ingredient information relating to fruit ingredients of the fruit-based food and drink; and a creation unit that creates recommendation criterion information for determining a recommendation level for the user with respect to a plurality of types of wine on the basis of the fruit preference information and the fruit ingredient information acquired by the acquisition unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a wine recommendation system and a wine recommendation method. [Background technology]

[0002] Patent Document 1 describes an algorithm that recommends appropriate wines to a user who requests preferences for each of wine characteristics such as price, grape variety, and taste. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2006-190126 A Summary of the Invention [Problem to be solved by the invention]

[0004] In Patent Document 1, wines are recommended to users who have specialized knowledge about wine grape varieties, tastes, etc. In other words, Patent Document 1 does not anticipate recommending wines to users such as beginners who do not have specialized knowledge about wine. An object of the present disclosure is to enable a user, even without expert knowledge of wine, to be recommended wines that suit the user's tastes. [Means for solving the problem]

[0005] A system according to one aspect of the present disclosure includes an acquisition unit that acquires fruit preference information regarding a user's preferences for fruit foods and beverages, and fruit component information regarding the fruit components of the fruit foods and beverages, and a creation unit that creates recommendation criteria information for determining a recommendation level for multiple types of wine to the user based on the fruit preference information and the fruit component information acquired by the acquisition unit.

[0006] A method according to one aspect of the present disclosure is a method executed by a system, and includes a step in which an acquisition unit provided in the wine recommendation system acquires fruit preference information regarding a user's preferences for fruit foods and beverages, and fruit component information regarding the fruit components of the fruit foods and beverages, and a step in which a creation unit provided in the wine recommendation system creates recommendation criteria information for determining a recommendation degree for multiple types of wine to the user based on the fruit preference information and the fruit component information acquired by the acquisition unit.

[0007] The embodiments of the present disclosure may be realized by an apparatus, a system, a method, an integrated circuit, a computer program, or a computer-readable non-transitory storage medium, or any combination thereof. The computer-readable storage medium may include both a volatile storage medium and a non-volatile storage medium. The apparatus may be composed of multiple separate devices. When composed of multiple separate devices, they may be arranged in one housing or may be arranged separately in two or more separate housings. Effect of the Invention

[0008] According to the present disclosure, even a user who does not have specialized knowledge of wine can be recommended wines that suit the user's tastes. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram showing the overall configuration of a wine recommendation system according to an embodiment. [Diagram 2] FIG. 2 is an explanatory diagram showing an example of an algorithm for creating recommendation criterion information, and shows normalization and vectorization of fruit components. [Diagram 3] FIG. 3 is an explanatory diagram showing an example of a creation algorithm of recommendation criterion information, and shows weighting of fruit component vectors. [Figure 4] FIG. 4 is an explanatory diagram showing an example of an algorithm for creating recommendation criterion information, and shows the summation of a plurality of fruit component vectors. [Diagram 5] FIG. 5 is an explanatory diagram showing an example of an algorithm for determining a recommendation level, and shows normalization and vectorization of wine components. [Figure 6] FIG. 6 is an explanatory diagram showing an example of an algorithm for determining the recommendation level, and shows how the recommendation level is determined from the cosine similarity. [Figure 7] FIG. 7 is a diagram showing an example of a display form of the recommendation level. [Figure 8] FIG. 8 is a diagram showing another example of a display form of the recommendation level. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] <Overview of the embodiment of the present disclosure> Below, an overview of the embodiments of the present disclosure will be listed and described. (1) The wine recommendation system of this embodiment includes an acquisition unit that acquires fruit preference information regarding a user's preferences for fruit foods and beverages, and fruit component information regarding the fruit components of the fruit foods and beverages, and a creation unit that creates recommendation criteria information for determining a recommendation level for multiple types of wine to the user based on the fruit preference information and the fruit component information acquired by the acquisition unit.

[0011] According to the wine recommendation system of the present embodiment, recommendation criteria information is created based on fruit preference information and fruit component information that indicate the user's preferences for fruit foods and beverages, and recommendation levels for multiple types of wine to the user are determined. Therefore, even if the user does not have specialized knowledge about wine, wines suitable for the user's preferences can be recommended based on the user's preferences for fruit foods and beverages.

[0012] (2) In the wine recommendation system of (1) above, it is preferable that the acquisition unit further acquires wine component information regarding the wine components of each of the plurality of types of wine, and further includes a determination unit that determines the recommendation level based on the recommendation criteria information and the wine component information. In this case, the recommendation level is determined using wine component information regarding the wine components of each of the multiple types of wine, so that wines that are more suited to the user's preferences can be recommended.

[0013] (3) In the wine recommendation system of (2) above, it is preferable that the determination unit determines the recommendation degree based on a correlation between the recommendation criterion information and the wine component information. In this case, a wine that is more suited to the user's tastes can be recommended.

[0014] (4) In the wine recommendation system of (2) above, the recommendation criteria information may include information regarding the fruit components that contain the same components as the wine components, and the determination unit may determine the recommendation degree by comparing the same components of the fruit components and the wine components. In this case, the recommendation level is determined by comparing the same fruit components and wine components, so that a wine that is more suited to the user's tastes can be recommended.

[0015] (5) In the wine recommendation system of (2) above, the recommendation criteria information may include information regarding the fruit components that contain the same components as the wine components, and the determination unit may determine the recommendation degree based on a correlation between the same components of the fruit components and the wine components. In this case, the recommendation level is determined based on the correlation between the same components of the fruit components and the wine components, so that a wine that is more suited to the user's preferences can be recommended.

[0016] (6) It is preferable that the wine recommendation system according to any one of (2) to (5) above further comprises a display unit that displays wine information regarding the plurality of types of wine in a form that allows the recommendation degree to be visually grasped. In this case, the user can easily understand the recommendation levels of multiple types of wine by looking at the display unit.

[0017] (7) The wine recommendation system according to any one of (2) to (5) above may further include a display unit that displays wine information about at least one wine among the plurality of wines in descending order of recommendation degree. In this case, the user can easily grasp information about the wine with the highest recommendation level from the wine information displayed on the display unit.

[0018] (8) In any of the wine recommendation systems described above in (1) to (7), it is preferable that the creation unit creates new recommendation criteria information when the fruit preference information is newly acquired by the acquisition unit. In this case, the recommendation level is determined using the most recent recommendation criterion information, so that wines that are more suited to the user's tastes can be recommended.

[0019] (9) In any of the wine recommendation systems described above in (2) to (8), it is preferable that the determination unit determines the recommendation degree anew when the wine component information is newly acquired by the acquisition unit or when the recommendation criteria information is newly created by the creation unit. In this case, the recommendation level is determined using the latest wine component information or the latest recommendation criterion information, so that a wine that is more suited to the user's tastes can be recommended.

[0020] (10) In any of the wine recommendation systems described above in (1) to (9), the fruit components are preferably at least one of sugar content, titratable acidity, malic acid, lactic acid, tartaric acid, hydrogen ion index (pH), tannins, and anthocyanins. In this case, the fruit component information is information about fruit components that are closely related to the taste of wine, so that a wine that is more suited to the user's preferences can be recommended.

[0021] (11) It is preferable that any of the wine recommendation systems described above in (1) to (10) further includes an input unit that accepts input of the fruit preference information, and the acquisition unit acquires the fruit preference information by an input operation on the input unit. In this case, since the fruit preference information is inputted by operating the input unit, the acquisition unit can easily acquire the fruit preference information.

[0022] (12) In any of the wine recommendation systems described above in (1) to (10), the acquisition unit may acquire the fruit preference information based on the fruit food or drink purchased by the user. In this case, for example, fruit preference information can be easily acquired based on the user's purchase history of fruit foods and beverages.

[0023] (13) In any of the wine recommendation systems described above in (1) to (12), it is preferable that the acquisition unit acquires the fruit preference information for each of a plurality of types of fruit food and beverages, and the fruit component information for each of the plurality of types of fruit food and beverages. In this case, the amount of fruit preference information and fruit component information increases, so that a wine that is more suited to the user's preferences can be recommended.

[0024] (14) The method according to this embodiment is a wine recommendation method executed in the wine recommendation system described above in (1) to (13). Therefore, the wine recommendation method of this embodiment has the same effects as the wine recommendation systems (1) to (13) described above.

[0025] <Details of the embodiment> Hereinafter, preferred embodiments will be described with reference to the drawings. [Overall system configuration] FIG. 1 is a block diagram showing the overall configuration of a wine recommendation system 1 according to an embodiment. The wine recommendation system 1 of this embodiment is composed of one terminal device 2. The terminal device 2 is, for example, a computer owned by a retailer that sells alcoholic beverages including wine. The terminal device 2 is used when a retailer recommends wines that suit the tastes of a target consumer (user) who is considering purchasing wine. The wine recommendation system 1 may be composed of multiple terminal devices.

[0026] The terminal device 2 includes a control unit 3, a storage unit 4, an input unit 5, and a display unit 6. The control unit 3 is an arithmetic processing device including a CPU (Central Processing Unit) and a RAM (Random Access Memory). The control unit 3 may include an integrated circuit such as an FPGA (Field-Programmable Gate Array). The control unit 3 reads a computer program 7 stored in the storage unit 4 into a main memory (RAM) and executes various information processing operations according to the read program 7.

[0027] The storage unit 4 is an auxiliary storage device including a non-volatile memory such as a hard disk drive (HDD) and a solid state drive (SSD). The storage unit 4 may include a flash read only memory (ROM), a universal serial bus (USB) memory, or an SD card. The programs 7 stored in the storage unit 4 include a program that causes the control unit 3 to execute a process of creating recommendation criteria information S4, which will be described later, a process of determining a recommendation level, and the like.

[0028] The input unit 5 is configured with an input interface such as a keyboard and a mouse, and accepts input of various information S1 to S3 (described later) by a user (such as a retailer or a target consumer) of the terminal device 2. The display unit 6 is configured with a display, for example, and displays various information. The input unit 5 and the display unit 6 may be integrated, for example, as a touch panel.

[0029] [Information registered in the memory unit] Fruit preference information S1, fruit component information S2, and wine component information S3 are preregistered in the storage unit 4. Therefore, the storage unit 4 of this embodiment corresponds to an acquisition unit that acquires the fruit preference information S1, the fruit component information S2, and the wine component information S3, respectively.

[0030] The fruit preference information S1 is information regarding the preference of the target consumer for each of a plurality of types of fruit food and beverages. The fruit food and beverages include fruits and processed fruit products (such as beverages and foods excluding wine). Examples of the processed products include fruit juice, fruit jelly, and dried fruit. The fruit preference information S1 may be information regarding the preference for a single fruit food and beverage.

[0031] The fruit preference information S1 of this embodiment is an evaluation value obtained by a target consumer evaluating his / her likes and dislikes for each of a plurality of types of fruit food and drink on a five-level scale from "1" to "5" as follows. The fruit preference information S1 is not limited to this embodiment, and may be an evaluation value evaluated on two or more levels. 5: I really like it 4: I like it 3: Normal 2: I don't like it 1: I really dislike it

[0032] The fruit preference information S1 is registered in the storage unit 4, for example, by a retailer conducting a questionnaire survey of target consumers regarding preferences for each of a plurality of types of fruit foods and beverages, and inputting the survey results into the input unit 5. The fruit preference information S1 may also be input into the input unit 5 by someone other than the retailer (for example, the target consumer).

[0033] The fruit preference information S1 may be created based on the fruit food and drink purchased by the target consumer. For example, the fruit preference information S1 may be created based on the target consumer's purchase history of multiple types of fruit food and drink, so that the more frequently purchased fruit food and drink, the higher the evaluation value. Examples of the purchase history include purchase history from real stores and purchase history from virtual stores on the web.

[0034] The fruit component information S2 is information on the fruit components for each of a plurality of types of fruit foods and beverages. The fruit components include the fruit components of the same types of fruit foods and beverages as those in the fruit preference information S1, but may also include the fruit components of other representative fruit foods and beverages. The fruit components are preferably at least one of sugar content, titratable acidity, malic acid, lactic acid, tartaric acid, hydrogen ion exponent (pH), tannins, and anthocyanins, which are closely related to the taste of wine. In this embodiment, the sugar content and titratable acidity are used as the fruit components, which are the same components as the wine components in the wine component information S3 described later. Hereinafter, titratable acidity is also simply referred to as acidity.

[0035] The fruit component information S2 is, for example, a value indicating the percentage of sugar content and acidity with respect to all components of each fruit food or drink. The percentages of sugar content and acidity are measured by a measuring device that non-destructively measures the fruit food or drink using, for example, spectroscopic analysis, and are registered in the storage unit 4 directly from the measuring device or via the input unit 5.

[0036] The wine component information S3 is information on the wine components of each of multiple types of wine. The type of wine includes the grape variety and the place of origin. Therefore, even if the grape variety is the same, if the place of origin is different, the wine will be of a different type. The wine components are preferably at least one of sugar content, titratable acidity, malic acid, lactic acid, tartaric acid, hydrogen ion exponent (pH), tannins, and anthocyanins. In this embodiment, sugar content and titratable acidity are used as the wine components.

[0037] The wine component information S3 is, for example, a value indicating the percentage of sugar content and acidity with respect to all components of each wine. The percentages of sugar content and acidity are measured, for example, by the above-mentioned measuring device, and are registered in the storage unit 4 directly from the measuring device or via the input unit 5.

[0038] [System control configuration] The control unit 3 has a creation unit 31 and a determination unit 32 as functional units of the program 7 executed by the control unit 3. After the various information S1 to S3 are registered in the storage unit 4 as described above, the creation unit 31 creates recommendation criteria information S4 based on the fruit preference information S1 and the fruit component information S2 in the storage unit 4. The recommendation criteria information S4 is information that serves as a criterion for determining the recommendation degree of multiple types of wine to a target consumer. Hereinafter, the recommendation degree of multiple types of wine to a target consumer is also simply referred to as recommendation degree.

[0039] The creation unit 31 passes the created recommendation criteria information S4 to the determination unit 32. The determination unit 32 determines the recommendation level based on the wine component information S3 in the storage unit 4 and the recommendation criteria information S4 created by the creation unit 31. The determination unit 32 of this embodiment determines the recommendation level based on the correlation between the wine component information S3 and the recommendation criteria information S4. The determination unit 32 may determine the recommendation level based on an index other than the correlation.

[0040] [Creating recommendation criteria information] 2 to 4 are explanatory diagrams showing an example of a creation algorithm of recommendation criteria information S4. Fig. 2 to Fig. 4 show an algorithm when recommendation criteria information S4 is created based on target consumer preferences and fruit components for three types of fruit foods and beverages (strawberries, mandarin oranges, and apples). Note that the three types of fruit foods and beverages are not limited to different fruits, and may be, for example, the same fruit of different varieties.

[0041] First, as shown in Fig. 2, the creation unit 31 normalizes the fruit component information S2 read from the storage unit 4 for each fruit component (sugar content and acidity), and creates a fruit component vector [sugar content, acidity] by vectorizing these normalized values ​​(step ST11). In this case, the creation unit 31 normalizes the highest value, which has the highest proportion of fruit components among multiple types of fruit food and drink, to "1", and normalizes the lowest value, which has the lowest proportion of fruit components, to "0". Furthermore, when the proportion of fruit components indicates an intermediate value between the maximum value and the minimum value, the creation unit 31 normalizes the intermediate value to a relative value between 0 and 1 using the following formula (1). Relative value = (median – minimum) / (maximum – minimum) (1)

[0042] In Figure 2, the creation unit 31 normalizes the sugar content ratio of the apple, which has the highest sugar content of the three types of fruit, "13.0", to "1". The creation unit 31 also normalizes the sugar content ratio of the strawberry, which has the lowest sugar content of the three types of fruit, "10.0", to "0". Furthermore, the creation unit 31 normalizes the sugar content ratio of the mandarin orange, which has the intermediate value, "11.0", to the relative value "0.33" using the above formula (1).

[0043] The creation unit 31 also normalizes the acidity ratio of each of the three types of fruits in the same manner as the sugar content ratio. Specifically, the creation unit 31 normalizes the acidity ratios of strawberries, mandarins, and apples, which are "1.0", "0.6", and "0.2", to "1", "0.5", and "0", respectively. Then, the creation unit 31 creates a fruit component vector by vectorizing the normalized sugar content ratio and the normalized acidity ratio for each of the three types of fruits. In FIG. 2, the fruit component vectors [sugar content, acidity] of strawberries, mandarins, and apples are respectively [0,1], [0.33,0.5], and [1,0].

[0044] 3, the creation unit 31 normalizes each evaluation value in the fruit preference information S1 between 0 and 1 as follows (step ST12). Hereinafter, the normalized evaluation value is also referred to as a normalized evaluation value. Evaluation value = 5 → Normalized evaluation value = 1.00 Evaluation value = 4 → Normalized evaluation value = 0.75 Evaluation value = 3 → Normalized evaluation value = 0.50 Evaluation value = 2 → Normalized evaluation value = 0.25 Evaluation value = 1 → Normalized evaluation value = 0.00

[0045] In Fig. 3, the evaluation values ​​of strawberries, mandarins, and apples are "4", "3", and "2", respectively. Therefore, the normalized evaluation values ​​of strawberries, mandarins, and apples are "0.75", "0.5", and "0.25", respectively. The normalization of the evaluation values ​​in step ST12 may be performed before the normalization or vectorization of the fruit components in step ST11.

[0046] Next, the creation unit 31 weights the fruit component vector of each fruit in the fruit component information S2 according to the normalized evaluation value of the corresponding fruit in the fruit preference information S1 (step ST13). In this embodiment, the creation unit 31 multiplies the fruit component vector of each fruit by the normalized evaluation value of the corresponding fruit. In FIG. 3, the fruit component vectors [sugar content, acidity] after weighting (multiplication) of strawberry, mandarin orange, and apple are [0, 0.75], [0.17, 0.25], and [0.25, 0], respectively.

[0047] Next, as shown in FIG. 4, the creation unit 31 sums up all the weighted fruit component vectors of each fruit (step ST14). In FIG. 4, the weighted fruit component vectors of strawberry, mandarin orange, and apple, respectively, [0, 0.75], [0.17, 0.25], and [0.25, 0] are all summed up to become [0.42, 1]. This summed fruit component vector becomes the recommendation reference information S4 of the target consumer. Hereinafter, the summed fruit component vector is also referred to as the summed fruit component vector.

[0048] After creating the recommendation criterion information S4, if the fruit preference information S1 is newly registered in the storage unit 4, the creation unit 31 creates new recommendation criterion information S4 by the procedure of steps ST11 to ST14 described above. For example, if the target consumer's likes and dislikes evaluation values ​​for fruits other than strawberries, mandarins, and apples are added to the fruit preference information S1, the creation unit 31 creates new recommendation criterion information S4 based on the new fruit preference information S1 and fruit component information S2.

[0049] In this case, the fruit component information S2 used is either the existing fruit component information S2 which already contains the fruit components of the fruit added to the fruit preference information S1, or if new fruit component information S2 containing the fruit components of the added fruit is registered in the memory unit 4, the new fruit component information S2 is used.

[0050] [Determination of recommendation level] 5 and 6 are explanatory diagrams showing an example of a recommendation level determination algorithm. Here, the algorithm is shown when the determination unit 32 determines recommendation levels for three types of wines A, B, and C using the recommendation criteria information S4 created by the above-mentioned procedure of steps ST11 to ST14.

[0051] 5, the determination unit 32 normalizes the wine component information S3 read from the storage unit 4 for each fruit component (sugar content and acidity), and creates a wine component vector [sugar content, acidity] by vectorizing these normalized values ​​(step ST21). The algorithm for creating the wine component vector is the same as that for the fruit component vector described above, so a detailed description will be omitted.

[0052] In FIG. 5, the determination unit 32 normalizes the sugar content ratio of wine C, which has the highest sugar content among the three types of wines A to C, from "3.8" to "1". The determination unit 32 also normalizes the sugar content ratio of wine A, which has the lowest sugar content among the three types of wines A to C, from "0.3" to "0". Furthermore, the determination unit 32 normalizes the sugar content ratio of wine B, which has the intermediate value, from "1.0", to the relative value "0.2" using the above-mentioned formula (1).

[0053] The determination unit 32 also normalizes the acidity ratio of each of the three types of wines A to C in the same way as the sugar content ratio. Specifically, the determination unit 32 normalizes the acidity ratio of wine A, which has the highest acidity value among the three types of wines A to C, from "0.7" to "1". The determination unit 32 also normalizes the acidity ratio of wine C, which has the lowest acidity value among the three types of wines A to C, from "0.5" to "0". Furthermore, the determination unit 32 normalizes the acidity ratio of wine B, which has the intermediate value, from "0.58", to "0.4", which is a relative value according to the above-mentioned formula (1).

[0054] Next, the determination unit 32 creates wine component vectors by vectorizing the normalized sugar content ratio and the normalized acid content ratio for each of the three types of wines A to C. In Fig. 5, the wine component vectors [sugar content, acidity] for wines A to C are [0,1], [0.2,0.4], and [1,0], respectively.

[0055] The determination unit 32 compares the combined fruit component vector with the wine component vector of each wine, that is, compares the same components (sugar content and acidity) of the fruit components and wine components, and finds the correlation between the two vectors (the same components). Here, as shown in FIG. 6, the determination unit 32 calculates the cosine similarity indicating the similarity between the combined fruit component vector and the wine component vector (step ST22). Specifically, the determination unit 32 calculates cosα, cosβ, and cosγ, which are the cosine similarities between the combined fruit component vector [0.42,1] and each of the wine component vectors [0,1], [0.2,0.4], and [1,0] of wines A, B, and C. The calculation results are as follows. Wine A: cosα ≒ 0.922 Wine B: cosβ ≒ 0.998 Wine C: cosγ≒0.387

[0056] The cosine similarity is an example of the correlation between both vectors (the recommendation criteria information S4 and the wine component information S3). This correlation may use a similarity other than the cosine similarity, or may use an index other than the similarity (for example, dissimilarity). The determination unit 32 may also compare both vectors after correcting the direction of the combined fruit component vector. For example, the determination unit 32 may perform a correction process such as multiplying the combined fruit component vector by a correction coefficient, and then compare the correction vector with the wine component vector.

[0057] Next, the determination unit 32 determines the recommendation degree for the multiple types of wines A to C to the target consumer based on the calculated cosine similarity (step ST23). The recommendation degree is, for example, a numerical value indicating the recommendation ranking, and the greater the cosine similarity, the smaller the value. Therefore, the smaller the value of the recommendation degree, the higher the recommendation degree to the target consumer. In FIG. 6, the wines B, A, and C are ranked in descending order of cosine similarity, so the determination unit 32 determines the recommendation degree of wine B to be "1", the recommendation degree of wine A to be "2", and the recommendation degree of wine C to be "3".

[0058] After creating the recommendation degree, if new wine component information S3 is registered in the storage unit 4, the determination unit 32 determines a new recommendation degree by the above-mentioned procedure of steps ST21 to ST23. For example, if wine components of a new type of wine are added to the wine component information S3, the determination unit 32 determines a new recommendation degree for the target consumer including the new type of wine.

[0059] In addition, even if the creation unit 31 creates new recommendation criteria information S4 after creating the recommendation degree, the determination unit 32 determines a new recommendation degree by the procedure of steps ST21 to ST23 described above. For example, when new fruit preference information S1 is registered as described above and the creation unit 31 creates new recommendation criteria information S4, the determination unit 32 uses the new recommendation criteria information S4 to determine a new recommendation degree for the target consumer.

[0060] [Display recommendation level] Fig. 7 is a diagram showing an example of a display form of the recommendation level. The display unit 6 displays the recommendation level determined by the determination unit 32 in a visually comprehensible form. As shown in Fig. 7, the display unit 6 simultaneously displays wine information on all three types of wines A to C in descending order of recommendation level. The wine information includes, for example, a photo of the wine bottle, product name, production area, grape variety, grape harvest year, volume, alcohol content, and taste, but may also include other information such as sugar content and acidity.

[0061] Fig. 8 is a diagram showing another example of the display form of the recommendation degree. As shown in Fig. 8, the display unit 6 displays only wine information about the wine with the highest recommendation degree. The display unit 6 may display the recommendation degree in a display form other than that shown in Fig. 7 and Fig. 8. For example, the display unit 6 may display wine information about all three types of wine, one screen at a time, in descending order of recommendation degree.

[0062] [Effects of the embodiment] As described above, the wine recommendation system 1 of the present embodiment creates recommendation criteria information S4 based on the fruit preference information S1 and fruit component information S2 indicating the target consumer's preference for fruit foods and beverages, and determines the recommendation level for multiple types of wine to the target consumer. Therefore, even if the target consumer does not have specialized knowledge of wine, it is possible to recommend a wine suitable for the target consumer's preference based on the target consumer's preference for fruit foods and beverages. In other words, since wine is recommended using the preference for "fruit" that has affinity with grapes, which are the raw material for wine, as an index, it is possible to determine a wine that is more preferred by the target consumer. In other words, when recommending a wine that is preferred by the target consumer from multiple types of wine, the present disclosure uses fruit as an index, and therefore can recommend a wine that is suitable for the target consumer's preference, compared to recommending wine using luxury items or foods other than fruit as an index.

[0063] The determination unit 32 determines the recommendation level using the recommendation criteria information S4 and the wine component information S3 regarding the wine components of each of the multiple types of wine, and is therefore able to recommend wines that are more suited to the preferences of the target consumer.

[0064] The decision unit 32 compares the combined fruit component vector, which is the recommendation criterion information S4, with the wine component vector, which is the wine component information S3, and calculates the similarity (cosine similarity) indicating the correlation between the two vectors. This makes it possible to recommend wines that are more suitable for the tastes of the target consumer.

[0065] The display unit 6 displays wine information on all three types of wine A to C side by side at the same time in a form that allows the recommendation level to be visually grasped. Therefore, by looking at the display unit 6, the target consumer can easily grasp the recommendation level of all three types of wine.

[0066] Furthermore, by displaying only the wine information about wine B, which has the highest recommendation level among the three types of wines A to C, the display unit 6 can easily grasp the wine information about wine B, which has the highest recommendation level.

[0067] The creation unit 31 creates new recommendation criterion information S4 when new fruit preference information S1 is registered in the storage unit 4. As a result, the recommendation level is determined using the latest recommendation criterion information S4, so that a wine that is more suitable for the taste of the target consumer can be recommended.

[0068] When new wine component information S3 is registered in the storage unit 4, or when the creation unit 31 creates new recommendation criteria information S4, the determination unit 32 determines a new recommendation degree. As a result, the recommendation degree is determined using the latest wine component information S3 or the latest recommendation criteria information S4, it is possible to recommend a wine that is more suitable for the tastes of the target consumer.

[0069] The fruit component information S2 is information about fruit components (sugar content and acidity) that are closely related to the taste of wine, so that it is possible to recommend a wine that is more suited to the tastes of the target consumer.

[0070] The fruit preference information S1 is input by a retailer or the like through the input unit 5, and therefore the fruit preference information S1 can be easily registered in the storage unit 4. In addition, even when the fruit preference information S1 is created based on fruit foods and beverages purchased by the target consumer, the fruit preference information S1 can be easily registered in the storage unit 4.

[0071] The fruit preference information S1 and the fruit component information S2 are information about the target consumer's preferences and information about fruit components for each of multiple types of fruit foods and beverages. Therefore, since the amount of information in the fruit preference information S1 and the fruit component information S2 is large, it is possible to recommend a wine that is more suitable for the target consumer's preferences.

[0072] [others] The embodiments disclosed herein are illustrative and not restrictive in all respects. The scope of the present invention is not limited to the above-described embodiments, but includes all modifications within the scope of the equivalents of the configurations described in the claims.

[0073] In the above embodiment, the fruit preference information S1, fruit component information S2, and wine component information S3 before normalization are registered in the memory unit 4, but the fruit preference information S1, fruit component information S2, and wine component information S3 after normalization may also be registered. The fruit component vector and the wine component vector are composed of two components (sugar content and acidity), but may be composed of one or three or more components. [Explanation of symbols]

[0074] 1. Wine recommendation system 2 Terminal Device 3. Control Unit 4 Storage section (acquisition section) 5 Input section 6 Display section 7. Computer Programs 31 Creation Department 32 Decision Section S1 Fruit preference information S2 Fruit composition information S3 Wine composition information S4 Recommendation Criteria Information

Claims

1. an acquisition unit that acquires fruit preference information related to a user's preference for fruit food or beverage, and fruit component information related to fruit components of the fruit food or beverage; A wine recommendation system comprising: a creation unit that creates recommendation criteria information for determining a recommendation degree for a plurality of types of wine to the user based on the fruit preference information and the fruit component information acquired by the acquisition unit.

2. The acquisition unit further acquires wine component information regarding wine components of each of the plurality of types of wine, The wine recommendation system according to claim 1 , further comprising a determination unit that determines the recommendation level based on the recommendation criterion information and the wine component information.

3. The wine recommendation system according to claim 2 , wherein the determination unit determines the recommendation degree based on a correlation between the recommendation criterion information and the wine component information.

4. the recommendation criteria information has information regarding the fruit component that contains the same component as the wine component; The wine recommendation system according to claim 2 , wherein the determination unit determines the recommendation degree by comparing the same components of the fruit components and the wine components with each other.

5. the recommendation criteria information has information regarding the fruit component that contains the same component as the wine component; The wine recommendation system according to claim 2 , wherein the determination unit determines the recommendation degree based on a correlation between the same components of the fruit components and the wine components.

6. The wine recommendation system according to claim 2 or 3, further comprising a display unit that displays wine information regarding the plurality of types of wine in a form that allows the recommendation degree to be visually grasped.

7. The wine recommendation system according to claim 2 or 3, further comprising a display unit that displays wine information on at least one wine from among the plurality of wines in descending order of the recommendation degree.

8. The wine recommendation system according to claim 1 , wherein the creation unit creates new recommendation criterion information when the fruit preference information is newly acquired by the acquisition unit.

9. 4. The wine recommendation system according to claim 2, wherein the determination unit determines the recommendation degree anew when the wine component information is newly acquired by the acquisition unit or when the recommendation criteria information is newly created by the creation unit.

10. The wine recommendation system of claim 1 or 2, wherein the fruit components are at least one of sugar content, titratable acidity, malic acid, lactic acid, tartaric acid, pH, tannins, and anthocyanins.

11. An input unit that receives the fruit preference information, The wine recommendation system according to claim 1 , wherein the acquisition unit acquires the fruit preference information through an input operation on the input unit.

12. The wine recommendation system according to claim 1 , wherein the acquisition unit acquires the fruit preference information based on the fruit food or drink purchased by the user.

13. The wine recommendation system according to claim 1 or 2, wherein the acquisition unit acquires the fruit preference information for each of a plurality of types of fruit food and drink, and the fruit component information for each of the plurality of types of fruit food and drink.

14. A method performed by a wine recommendation system, comprising: An acquisition unit of the wine recommendation system acquires fruit preference information related to a user's preference for fruit food or beverage, and fruit component information related to fruit components of the fruit food or beverage; a step of a creation unit included in the wine recommendation system creating recommendation criteria information for determining a recommendation level for a plurality of types of wine to the user based on the fruit preference information and the fruit component information acquired by the acquisition unit.

Images