KR20240021017A - Wine recommendation system using wine style matching and method thereof - Google Patents

Abstract

The present invention relates to a wine recommendation system and method using wine style matching. The wine recommendation system of the present invention includes a wine recommendation server that provides a wine subscription service that recommends wine based on customer surveys and recommends the next wine purchase based on feedback on the recommended wine. The wine recommendation server defines and classifies various types of wine in advance into multiple wine styles with standard values based on wine attributes including at least body, acidity, tannin, and sweetness, and uses artificial intelligence to provide customized wines to customers. We provide a recommended wine subscription service, through which we derive the customer's preferred wine style based on customer surveys and match the derived wine style with a predefined wine style to recommend wine to the customer. According to the present invention, by recommending wines that match the customer's preferred wine style through surveys and feedback using a regular wine subscription service, the customer can save additional time and money for selecting wine even if the customer does not have knowledge about wine. It has the effect of allowing you to easily select and enjoy wine without spending any money.

Description

Wine recommendation system and method using wine style matching {WINE RECOMMENDATION SYSTEM USING WINE STYLE MATCHING AND METHOD THEREOF}

The present invention relates to a wine recommendation system, and more specifically, to define and classify various types of wine in advance into a plurality of wine styles with reference values based on wine attributes including at least body, acidity, tannin, and sweetness, We provide a wine subscription service that uses artificial intelligence to recommend customized wines to customers. Through this, we derive the customer's preferred wine style based on the first customer's survey and match the derived wine style with the defined wine style. It relates to a wine recommendation system and method using wine style matching that recommends wine to a customer, receives feedback on wine attribute settings, evaluation, and satisfaction level for the recommended wine, and processes it to recommend the next wine.

In the recent wellness era, interest in and demand for wine is explosively increasing. In response to this, wine marketing methods, such as providing consumers with a wider range of choices and performing efficient inventory management, are becoming very important tasks in the wine industry.

Wine is no longer an alcoholic beverage consumed depending on a special place, time, or situation, but is now recognized by consumers as an alcoholic beverage that can be easily and easily enjoyed by the public anytime, anywhere. The types of these wines are very diverse compared to other alcoholic beverages. Currently, there are over 50,000 types of wine imported into the country, and the average wine type that a store that sells wine usually stocks is over 400 types. It is said that it exceeds. This is because wine is a product of preference that is extremely dependent on individual subjectivity and taste, and each type of wine has different specifications depending on the country of production, winery, vintage, type, variety, sweetness, acidity, body, food that goes well with it, etc. Therefore, excessive inventory management of various types of wine is a major cause of worsening profits in the wine industry.

Meanwhile, from the consumer's perspective, unless they are an expert, it is difficult to select a wine that suits their taste or is preferred due to the lack of knowledge about wine and the variety of wine types. In particular, consumers who encounter wine for the first time have limited knowledge about wine. There are none, so choosing a wine is very difficult. Therefore, with the perception that wine is difficult, when actually purchasing wine, many consumers have difficulty choosing wine because they are vague about what criteria to use among the numerous wines.

Korean Patent Publication No. 10-2057083 (announcement date: December 18, 2019) Korean Patent Publication No. 10-2087824 (announcement date April 23, 2020) Korean Patent Publication No. 10-1843987 (announcement date: March 30, 2018) Korean Patent Publication No. 10-2079289 (announcement date April 7, 2020)

The purpose of the present invention is to recommend a wine based on a customer survey in order to provide a customized wine to the customer, and a wine recommendation system using wine style matching that recommends the wine based on the customer's feedback on the recommended wine, and the same. It provides a method.

Another object of the present invention is to provide a wine recommendation system and method using wine style matching that recommends customized wines to customers using artificial intelligence and provides regular wine subscription services for wine recommendations and purchases to customers. .

Another object of the present invention is to define and classify various types of wine in advance into a plurality of wine styles with reference values based on a plurality of wine attributes, to derive the wine style preferred by the customer based on the customer's survey, The goal is to provide a wine recommendation system and method using wine style matching that recommends wine to customers by matching the derived wine style with a defined wine style.

In order to achieve the above objectives, the wine recommendation system of the present invention defines and classifies various types of wine in advance into a plurality of wine styles with reference values based on a plurality of wine attributes, and determines the customer's preference based on the customer's questionnaire. We derive a wine style, match the derived wine style with a defined wine style, recommend a customized wine to the customer, and receive feedback on the attributes, evaluation, and satisfaction of the recommended wine from the customer to select the next wine for the customer. One feature is that it provides a regular subscription service for recommended wines.

The wine recommendation system of the present invention matches the wine style and recommends a wine suitable for the customer, and as the number of subscriptions to the regular wine subscription service passes, it recommends a customized wine suitable for the customer's taste, allowing the customer to easily select a wine. Let it happen.

The wine recommendation system of the present invention according to this feature includes a wine recommendation server that provides a wine subscription service that allows customer terminals of customers registered as members through a communication network to purchase recommended wines by regularly subscribing to wine online; Including, the wine recommendation server sets classification values for each wine category of a plurality of wines and attribute values for a plurality of wine attributes, and defines the wine style for each of the wines as a standard value according to the wine attributes. a wine information management unit that classifies and creates a wine style list containing defined wine styles; When registering as a member of the customer terminal, a customer management unit that provides different types of wine subscription services according to membership level; a payment processing unit that processes payment for a wine subscription service according to the membership level of the customer terminal by linking the customer terminal with an external payment system; When first registering as a member of the customer terminal, survey information about the wine category and wine attributes is provided to the customer terminal, and survey result data corresponding to the survey information is received from the customer terminal to derive the customer's preferred wine style and A survey processing unit that classifies and generates the derived and classified wine styles as a customer's wine style list; Provide feedback survey information about the wine recommended to the customer terminal, receive feedback data corresponding to the feedback survey information from the customer terminal, reclassify the customer's wine style, and convert the reclassified wine style into the customer's wine style. a feedback processing unit that updates the list; Compare the wine style list generated from the survey processing unit with the wine style list preset from the wine information management unit and recommend at least one wine corresponding to a predefined wine style that matches the customer's preferred wine style to the customer terminal. and a wine recommendation processing unit that compares the wine style list updated from the feedback processing unit with a preset wine style list and recommends the next wine to the customer terminal. and a control unit that controls the wine information management unit, the customer management unit, the payment processing unit, the survey processing unit, the feedback processing unit, and the wine recommendation processing unit to process operations of the wine recommendation server in conjunction with the customer terminal and the payment system. Includes.

In this feature, the survey processing unit extracts wine attribute values from survey result data, applies preset weights to the extracted wine attribute values, calculates survey values for the customer's wine attributes, and calculates survey values for the wine attributes. Analyzing the wine attribute style according to the customer's wine attribute taste by calculating the average of the survey values of the wine attribute by applying the deviation from the wine attribute value of the preset wine style list; Wine category classification values are extracted from the survey result data, and a preset weight is applied to the extracted wine category classification values to analyze the wine category style according to the customer's wine category taste; Next, the customer's wine style list is classified and created using the analyzed wine attribute style and wine category style.

In this feature, the feedback processing unit extracts wine attribute values from the feedback data, applies preset weights to the extracted wine attribute values, recalculates the customer's survey value for the wine attribute, and recalculates the survey value for the wine attribute. The average of the wine attribute survey values is calculated by applying the deviation from the wine attribute values of the preset wine style list, and the customer's wine attribute style is reanalyzed by applying weights according to time to the wine attribute survey values. Update the wine style list.

In this feature, the wine recommendation processing unit analyzes survey result data or feedback data using artificial intelligence to generate or update a wine recommendation list corresponding to the customer's wine style list; One wine is automatically selected and recommended from the created or updated wine recommendation list and transmitted to the customer terminal, or a wine recommendation list is transmitted to the customer terminal to select a wine.

In this feature, the wine information management unit classifies wines into red wines, white wines, and sparkling wines according to various types of wine, and wines that contain at least the body, acidity, tannin, and sweetness of wines, each classified into 5 levels. According to the standard values of the attributes, red wines are defined and classified into first to third red wine styles, white wines are defined as first to third white wine styles, and sparkling wines are defined and classified as first to third sparkling wine styles; The first red wine style is defined as wines with standard values of wine properties such as body grade 4 or higher, acidity grade 4 or higher, tannin grade 4 or higher, and sweetness ignored, and at the same time, body, acidity, and tannin. This also applies to wines where one of the reference values is 3 stars but the other two are 5 stars; The second red wine style is defined as wines that have standard values of wine properties such as body grade 3 or higher, acidity grade 3 or higher, tannin grade 3 or higher, and sweetness ignored, and at the same time, body, acidity, and tannin. It does not apply to wines in which any one of the standard values is below grade 2, excludes wines in which both body and tannin are grade 3, and applies to all wines excluding the first red wine style among wines that satisfy all of the above conditions. become; The third red wine style is defined as wines that have body of 3 or less grades, acidity of 4 or more, tannin of 3 or less, and sweetness ignored as standard values for wine properties, and at the same time, body, acidity and tannin are defined as standard values. Applies to wines that meet all criteria; The first white wine style is defined as wines having a body of 3 or more grades, acidity of 4 or more, and tannin and sweetness ignored as standard values for wine properties; The second white wine style is defined as wines with standard values of wine properties: body of 3 or higher, acidity of 3 or lower, and tannin and sweetness ignored; The third white wine style is defined as wines with standard values of wine properties: body of 2 grades or less, acidity of 4 grades or more, and tannin and sweetness ignored; The first sparkling wine style is defined as wines that have a sweetness level of 3 or higher as a standard value of wine attributes and the standard values of other wine attributes are ignored; And the second sparkling wine style is defined as wines in which the sweetness level is 2 or less as a standard value for wine properties and the standard values for other wine properties are ignored.

As described above, the wine recommendation system of the present invention recommends wine based on a customer survey, receives feedback on the recommended wine, and recommends wine, allowing the customer to select a wine even if he or she has no knowledge about wine. It has the effect of allowing you to easily select and enjoy wine without spending extra time or money.

In addition, the wine recommendation system of the present invention provides a wine subscription service that recommends wine based on customer surveys and feedback and recommends wine on a regular basis. As the user's preference is reflected, increasingly accurate taste analysis is possible and wine recommendation is made. Accuracy can be improved, and as the number of subscriptions passes, customized wines suitable for the customer's taste are recommended, allowing the customer to easily select a wine.

In addition, the wine recommendation system of the present invention recommends wine to customers by matching the customer's preferred wine style with a predefined wine style, so that customers do not have difficulty purchasing wine, thereby improving satisfaction and providing service. Wine sales at wineries increase, and customized wines can be provided to customers. Accurate wine sales trends can be identified, and accurate wine marketing strategies can be established, making wine inventory management easier.

In addition, the wine recommendation system of the present invention provides a wine subscription service using artificial intelligence, processing the entire process from customer surveys to ordering, packaging and delivery, allowing customers to easily purchase wine and experience wine culture.

In addition, the wine recommendation system of the present invention provides a customized wine recommendation service by identifying the customer's taste, budget, and needs, thereby enabling reasonable and optimal wine selection, and enabling customers to select a wine through an online platform or consultation service. You can receive a variety of information about your favorite wine, providing enjoyment through a wine subscription.

In addition, the wine recommendation system of the present invention can establish special marketing by allowing wine to be gifted to others using a wine subscription service.

1 is a block diagram showing the network configuration of a wine recommendation system using wine style matching according to the present invention;
Figure 2 is a block diagram showing the configuration of the wine recommendation server shown in Figure 1;
Figure 3 is a flowchart showing the wine recommendation procedure of the wine recommendation system using wine style matching according to the present invention;
Figure 4 is a flow chart showing the procedure for recommending wine based on the wine questionnaire from the first customer shown in Figure 3, and
FIG. 5 is a flowchart showing a procedure for recommending wine based on feedback data about the recommended wine from the customer shown in FIG. 3 .

Embodiments of the present invention may be modified in various forms, and the scope of the present invention should not be construed as being limited by the embodiments described below. This example is provided to more completely explain the present invention to those with average knowledge in the art. Therefore, the shapes of components in the drawings are exaggerated to emphasize a clearer explanation.

The wine recommendation system using wine style matching according to the present invention determines wine styles for various types of wine in advance, in order to provide a customized wine subscription service, and wine attributes including body, acidity, tannin, and sweetness. It defines and classifies based on standard values, provides customers with survey information on preferred wine categories and wine attributes, receives survey result data corresponding to the survey information, and analyzes and analyzes the customer's wine style using artificial intelligence. Classifies and compares the customer's wine style derived from the survey with the preset wine style to create a wine recommendation list suitable for the customer's taste, and automatically selects at least one wine that matches the survey result data from the wine recommendation list. So we recommend it to our customers.

In addition, the wine recommendation system according to the present invention receives feedback data by providing feedback survey information about the recommended wine, and analyzes the wine attributes, evaluation, and satisfaction of the recommended wine using artificial intelligence to determine the customer's wine style. are reclassified, and through this, wines for the next round are recommended.

Therefore, the wine recommendation system according to the present invention provides a regular wine subscription service using artificial intelligence, and handles the entire process from customer surveys to ordering, packaging and delivery, allowing customers to easily purchase wine and experience wine culture.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the attached FIGS. 1 to 5.

Figure 1 is a block diagram showing the network configuration of a wine recommendation system using wine style matching according to the present invention, and Figure 2 is a block diagram showing the configuration of the wine recommendation server shown in Figure 1.

Referring to Figures 1 and 2, the wine recommendation system 2 using wine style matching according to the present invention provides a wine recommendation service for regular wine subscription to customers, that is, a wine subscription service, in order to recommend customized wines to customers. Provides subscription service.

To this end, the wine recommendation system (2) of the present invention defines a plurality of wine styles in advance according to wine attributes for various wines, classifies the defined wine styles according to the standard values of wine attributes, and informs the first customer of the wine style. By providing survey information about this preferred wine, we receive survey result data corresponding to the survey information from the customer, and use artificial intelligence (AI) based on the customer's survey result data to recommend wine that matches the preset wine style. Then, feedback survey information about the recommended wine is provided to customers who have subscribed to the recommended wine, feedback data corresponding to the feedback survey information is received from the customer, and the next wine is recommended based on the feedback data.

Specifically, the wine recommendation system 2 of the present invention defines and classifies a plurality of wine styles using characteristics according to wine attributes. Here, wine styles are classified into red wine, white wine, and sparkling wine depending on the type of wine, with red wine having first to third red wine styles, white wine having first to third white wine styles, and sparkling wine having first to third red wine styles. It is classified by definition into the first to first sparkling wine styles.

In the present invention, the first red wine style is defined as a red wine with a smooth body and strong acidity and tannin, and the second red wine style is defined as a red wine with a heavy or medium body and moderate acidity and tannin. The third red wine style is defined as a red wine with a relatively light body, strong acidity, and low tannin.

In other words, the first red wine style is defined as wines that have body of 4 or more, acidity of 4 or more, tannin of 4 or more, and sugar content of 4 or more as standard values for wine properties. At this time, the first red wine style also applies to wines where one of the above three standard values is 3, but the other two are 5. The second red wine style is defined as wines with a body of 3 or more, acidity of 3 or more, tannin of 3 or more, and sweetness ignored. At this time, the second red wine style does not apply to wines that have any of the above three standard values below 2. In addition, the second red wine style is excluded for wines where both body and tannin are 3, and in this case, the third red wine style is applied. Additionally, the second red wine style applies to all wines that satisfy all of the above conditions except the first red wine style. The third red wine style is defined as wines with body of 3 or less, acidity of 4 or more, tannin of 3 or less, and sweetness ignored. At this time, the third red wine style is applied to wines that satisfy all three conditions above.

The first white wine style is defined as a white wine with a heavy to medium body and strong acidity, the second white wine style is defined as a white wine with a heavy to medium body and a moderate acidity, and the third white wine style is defined as a white wine with a relatively light body and strong acidity.

In other words, the first white wine style is defined as wines with a body of 3 or more, acidity of 4 or more, and tannin and sweetness ignored. The second white wine style is defined as wines with a body of 3 or more, acidity of 3 or less, and tannin and sweetness ignored. The third white wine style is defined as wines with body of 2 or less, acidity of 4 or more, and tannin and sweetness ignored.

Additionally, the first sparkling wine style is defined as a sweet sparkling wine with high sugar content, and the second sparkling wine style is defined as a brut sparkling wine with low sugar content.

That is, the first sparkling wine style is defined as wines with a sugar content of 3 or more and the standard values of other wine properties are ignored. The second sparkling wine style is defined as wines with a sugar content of 2 or less and reference values for other wine attributes ignored.

The wine recommendation system 2 of the present invention derives the customer's preferred wine style through the customer's survey results, matches the derived wine style with a predefined wine style, and recommends a customized wine suitable for the customer.

In addition, the wine subscription service of the present invention recommends wine to subscribe to wine on a regular basis (for example, once or twice a month) and provides an online service so that the customer can purchase at least one recommended wine. At this time, the wine subscription service recommends wine to the customer and provides various information about the recommended wine, such as producer, product name, variety, country of origin, vintage, grade, alcohol content, type, and drinking temperature. , provides additional wine-related content such as decanting, food, evaluation, price, and reviews.

Therefore, the wine recommendation system 2 of the present invention uses artificial intelligence to regularly recommend customized wines to customers, allowing customers to easily select and access wine.

To this end, the wine recommendation system 2 of the present invention includes a customer terminal 300 and a wine recommendation server 100. In addition, the wine recommendation system 2 of the present invention further includes a payment system 200. The customer terminal 300, the wine recommendation server 100, and the payment system 200 are connected to each other through a communication network 10.

Specifically, the communication network 10 includes, for example, a wired and wireless communication network, a mobile communication network, etc., and may be provided as a single communication network or a combined complex communication network. The communication network 10 connects a plurality of customer terminals 300, the wine recommendation server 100, and the payment system 200 to enable mutual data communication.

The customer terminal 300 is equipped with a communication capable electronic device, such as a personal computer, tablet PC, smart phone, tablet phone, etc., and is connected to the wine recommendation server 100 through the communication network 10. The customer terminal 300 registers as a member by entering the customer's personal information (eg, name, contact information, ID, password, email address, SNS information, etc.) into the wine recommendation server 100. At this time, the customer terminal 300 determines the membership level of the wine subscription service and signs up, pays a fee corresponding to the determined membership level on a monthly basis, and according to the paid fee, at least one wine suitable for the membership level is purchased. are regularly recommended.

When the customer terminal 300 first registers as a member, the customer terminal 300 receives survey information about the customer's preferred wine from the wine recommendation server 100. In an embodiment of the present invention, the survey information includes a survey of preferences for wine categories such as the type of wine the customer prefers or dislikes, grape variety, country of production, and production region, and wine properties such as body, acidity, sweetness, and tannin. Content is included. The wine category is set by selection or direct input, and the wine attribute is set by entering one of a plurality of levels, for example, 5 levels, to set preference.

The customer terminal 300 responds to the survey information and inputs and transmits the survey result data to the wine recommendation server 100. The customer terminal 300 receives at least one wine recommendation according to the customer's preferred wine style from the wine recommendation server 100 in response to the survey result data. Of course, the customer terminal 300 may change the recommended wine by re-responding to the survey information or modifying the survey result data. Accordingly, customers can have wine recommended by the wine subscription service delivered to their desired location.

Additionally, when the customer terminal 300 subscribes to a wine recommended by the wine recommendation server 100, it is provided with feedback survey information that surveys the wine attributes, evaluation, satisfaction, etc. for the recommended wine. Feedback survey information also includes a survey on the wine category, wine attributes, and satisfaction with the recommended wine. The customer terminal 300 inputs and transmits feedback data to the wine recommendation server 100 in response to the feedback survey information. The customer terminal 300 receives a wine recommendation corresponding to the feedback data from the wine recommendation server 100 so that it can be subscribed to in the next round.

The payment system 200 is linked to the wine recommendation server 100 and the customer terminal 300 to process the customer's wine subscription service application and payment of fees according to the membership level. The payment system 200 processes payments on a monthly basis according to the membership level of the customer terminal 300.

The payment system 200 is equipped to enable the customer terminal 300 to make payments using various payment methods, such as credit cards, check cards, app cards, bank transfers, and online simple payment services. This payment system 200 is a typical payment processing system for online payment processes such as payment request, approval, and payment processing. Since various technologies for this payment system 200 are already known, detailed descriptions are omitted here.

And the wine recommendation server 100 provides a wine subscription service using artificial intelligence according to the present invention to a plurality of customer terminals 300. The wine recommendation server 100 recommends wine to customers based on a survey through a regular wine subscription service, and receives feedback on the recommended wine from customers who have subscribed to the recommended wine to provide wine more suitable to the customer's taste. I recommend you.

To this end, the wine recommendation server 100 creates and stores a wine style list in which wine categories and wine attributes for each of a plurality of wines are set in advance. Here, the wine style list sets classification values for wine categories such as red, white, still, sparkling, rosé, and fortified wine, and attribute values for wine attributes such as body, acidity, sweetness, and tannin for each wine. . At this time, wine attribute values are set to multiple levels (for example, 5 levels). In other words, the body is set to five grades (standard values of grades 1 to 5) according to light and heavy, the acidity is set to five grades (standard values of grades 1 to 5) according to low and high, and the sweetness is set to low and high. It is set to five grades (standard values of grades 1 to 5), and tannin is set to five grades (standard values of grades 1 to 5) depending on dryness and sweetness.

In an embodiment of the present invention, the wine recommendation server 100 defines and classifies various types of wine in advance into a plurality of wine styles with reference values based on wine attributes including at least body, acidity, tannin, and sweetness. In other words, wine styles are classified into red wine, white wine, and sparkling wine depending on the type of wine, with red wine having first to third red wine styles, white wine having first to third white wine styles, and sparkling wine having first to third red wine styles. It is classified and defined into the first to first sparkling wine styles. At this time, the wine style is defined and classified through standard values based on wine attributes including at least body, acidity, tannin, and sweetness of the wine.

Here, the first red wine style is defined as a red wine with a solid body and strong acidity and tannin, and the second red wine style is defined as a red wine with a heavy or medium body and moderate acidity and tannin, The third red wine style is defined as a red wine with a relatively light body, strong acidity, and low tannin.

In other words, the first red wine style is defined as wines that have body of 4 or more, acidity of 4 or more, tannin of 4 or more, and sugar content of 4 or more as standard values for wine properties. At this time, the first red wine style also applies to wines where one of the above three standard values is 3, but the other two are 5. The second red wine style is defined as wines with a body of 3 or more, acidity of 3 or more, tannin of 3 or more, and sweetness ignored. At this time, the second red wine style does not apply to wines that have any of the above three standard values below 2. In addition, the second red wine style is excluded for wines where both body and tannin are 3, and in this case, the third red wine style is applied. Additionally, the second red wine style applies to all wines that satisfy all of the above conditions except the first red wine style. The third red wine style is defined as wines with body of 3 or less, acidity of 4 or more, tannin of 3 or less, and sweetness ignored. At this time, the third red wine style is applied to wines that satisfy all three conditions above.

The first white wine style is defined as a white wine with a heavy to medium body and strong acidity, the second white wine style is defined as a white wine with a heavy to medium body and a moderate acidity, and the third white wine style is defined as a white wine with a relatively light body and strong acidity.

In other words, the first white wine style is defined as wines with a body of 3 or more, acidity of 4 or more, and tannin and sweetness ignored. The second white wine style is defined as wines with a body of 3 or more, acidity of 3 or less, and tannin and sweetness ignored. The third white wine style is defined as wines with body of 2 or less, acidity of 4 or more, and tannin and sweetness ignored.

Additionally, the first sparkling wine style is defined as a sweet sparkling wine with high sugar content, and the second sparkling wine style is defined as a brut sparkling wine with low sugar content.

That is, the first sparkling wine style is defined as wines with a sugar content of 3 or more and the standard values of other wine properties are ignored. The second sparkling wine style is defined as wines with a sugar content of 2 or less and reference values for other wine attributes ignored.

The wine recommendation server 100 registers the customer of the customer terminal 300 as a member. At this time, the wine recommendation server 100 determines the membership level by providing guidance information on the membership level so that the customer terminal 300 can select the desired membership level.

The wine recommendation server 100 provides different types of wine subscription services according to the membership level applied for by the customer terminal 300. For example, the membership level is provided to the customer terminal 300 by varying the number of bottles of wine that can be purchased on a regular basis, the price range of the wine, the quality level of the wine, the number of purchases, the availability of gifts, etc., and the wine recommendation server 100 provides the corresponding type of wine subscription service to the customer terminal 300 according to the membership level.

When a customer is registered as a member, the wine recommendation server 100 provides survey information to the customer terminal 300 to survey the customer's wine preference for the first time. The wine recommendation server 100 receives survey result data corresponding to the survey information from the customer terminal 300, analyzes the wine category and wine attributes of the customer's taste using artificial intelligence, and derives the customer's preferred wine style. A wine recommendation list containing at least one wine is created and stored by classifying, comparing and matching a wine style list containing a plurality of preset wine styles. The wine recommendation server 100 automatically selects at least one wine using artificial intelligence from the generated wine style list and recommends it to the customer terminal 300. Of course, the wine recommendation server 100 may transmit a wine recommendation list to the customer terminal 300 so that the customer terminal 300 can select a wine.

In addition, the wine recommendation server 100 recommends wine to the customer terminal 300 and then transmits feedback survey information to the customer terminal 300 to survey wine attributes, evaluation, satisfaction, etc. for the recommended wine. At this time, the wine recommendation server 100 notifies the customer terminal 300 that feedback survey information about the recommended wine has been sent to the customer terminal 300 using a text message or the like. The feedback survey information is provided with largely the same or similar content as the survey information, and includes wine attribute values for wines recommended in the first or previous round, and surveys on customer evaluation and satisfaction.

In addition, the wine recommendation server 100 receives feedback data corresponding to the feedback survey information from the customer terminal 300, analyzes the customer's preferred wine using artificial intelligence, reclassifies the wine style, and uses the feedback data and preset Compare and match wine style lists to update the customer's wine recommendation list. The wine recommendation server 100 automatically selects and recommends wines using artificial intelligence from the wine recommendation list updated to the customer terminal 300 and provides the customer terminal 300 to subscribe to the recommended wine in the next round, or provides the updated wine recommendation server to the customer terminal 300. A wine recommendation list is transmitted so that the customer terminal 300 selects a wine. Therefore, the wine recommendation server 100 can improve the accuracy of recommending wine suitable for the customer's taste based on feedback data by repeating the process of the wine subscription service.

This wine recommendation server 100 provides recommended wine and wine-related content to customers, including an online shopping mall function from wine selection (i.e., ordering) to courier delivery, along with a regular wine subscription service.

Specifically, the wine recommendation server 100 of the present invention includes a server unit 102 and a database unit 104. The server unit 102 is, for example, a web server and app server that processes a plurality of customer terminals 300 and the payment system 200 to enable data communication through the communication network 10, and a wine implemented based on artificial intelligence. The hardware and software platform for the subscription service is equipped with an integrated application server. The database unit 104 stores and manages various information according to the overall processing process of the wine recommendation server 100 processed by the server unit 102. In this embodiment, the database unit 104 is included in the wine recommendation server 100, but may also be provided as an independent database server.

The server unit 102 in this embodiment includes a control unit 110, a communication unit 112, a wine information management unit 114, a customer management unit 116, a survey processing unit 118, a feedback processing unit 120, and a wine recommendation processing unit 122. , including a payment processing unit 124 and a consultation processing unit 126.

The control unit 110 controls the overall operations of the wine recommendation server 100 in conjunction with the customer terminal 300 and the payment system 200. That is, the control unit 110 includes a communication unit 112, a wine information management unit 114, a customer management unit 116, a survey processing unit 118, a feedback processing unit 120, a wine recommendation processing unit 122, a payment processing unit 124, The functions of the consultation processing unit 126 and the database unit 104 are controlled so that they are processed organically.

The communication unit 112 is provided with, for example, a wired or wireless communication module, a mobile communication module, etc., and is connected to each of a plurality of customer terminals 300 and the payment system 200 through the communication network 10 to enable mutual data communication. Process it.

The wine information management unit 114 stores and manages various information about a plurality of wines in a database. The wine information management department 114 classifies various types of wine into red wine, white wine, and sparkling wine in advance, and wine properties that include at least body, acidity, tannin, and sweetness of wine, each classified into 5 grades. Depending on the reference value (i.e., rating value for wine attributes), red wines are defined as first to third red wine styles, white wines are defined as first to third white wine styles, and sparkling wines are defined as first to third sparkling wine styles. and classify it.

That is, the wine information management unit 114 registers and manages wine information, wine category information, and wine attribute information for each of a plurality of wines. Wine information includes, for example, wine name (product name), wine variety, country of production, production area, place of origin, vintage, grade, alcohol content, drinking temperature, food compatibility, etc.

Wine category information includes wine types, such as red, white, still, sparkling, and rosé. And wine attribute information includes, for example, body, acidity, sweetness, and tannin. Wine attribute information is set into a plurality of grades (e.g., 1 to 5 grades) according to good/bad, high/low, high/low, etc. for each of body, acidity, sweetness, and tannin. The wine information management unit 114 sets classification values for wine category information and attribute values for wine attribute information for each wine, and uses these to classify predefined wine styles for a plurality of wines to create a wine style. Create and save a list. Additionally, the wine information management unit 114 matches and stores various information about each wine, that is, wine-related content, to each wine. Wine information, wine style list, and wine-related contents generated by the wine information management unit 114 are stored and managed in the database unit 104.

The customer management unit 116 receives the customer's personal information from each of the plurality of customer terminals 300 and registers and manages the member information in the database unit 104. The customer management unit 116 sets and manages member information to provide different types of wine subscription services according to the membership level of each of the customer terminals 300. The customer management unit 116 stores the customer's wine style by including it in the member information in response to the survey result data of each customer terminal 300. The customer management unit 116 matches, stores, and manages the customer's wine recommendation list corresponding to the member information and feedback data.

When a customer first signs up as a member, the survey processing unit 118 transmits survey information to the customer terminal 300 and receives survey result data in response to the survey information from the customer terminal 300. The survey processing unit 118 analyzes the customer's preferred wine category classification value and wine attribute value using artificial intelligence from the survey result data to derive and classify the customer's wine style. The survey processing unit 118 stores and manages the customer's wine style in member information.

The feedback processing unit 120 transmits feedback survey information that surveys wine attributes, evaluation, satisfaction, etc. for the recommended wine to the customer terminal 300 that has subscribed to the wine, and responds to the feedback survey information from the customer terminal 300. Receive feedback data. At this time, the feedback processing unit 120 notifies the customer terminal 300 through a text message or the like that feedback survey information has been transmitted. The feedback processing unit 120 analyzes feedback data using artificial intelligence, reclassifies the customer's wine style, and updates the wine style in the customer's member information.

The wine recommendation processing unit 122 analyzes the survey result data from the survey processing unit 118 and the feedback data from the feedback processing unit 120 to generate and store a wine recommendation list that matches the customer's wine style. That is, the wine recommendation processing unit 122 compares the customer's wine style classified from the survey result data with the wine style list preset in the database unit 104 to include at least one wine with a wine style that matches the customer's wine style. Create and save a recommended wine list. The wine recommendation processing unit 122 compares the customer's wine style updated from the feedback data with the wine style list and updates the wine recommendation list including wines matching the customer's wine style.

The wine recommendation processing unit 122 uses artificial intelligence to automatically select and recommend one wine from the wine recommendation list corresponding to the survey result data or feedback data and transmits it to the customer terminal 300, or sends the wine recommendation list to the customer terminal. Send it to (300) and select wine.

The payment processing unit 124 works in conjunction with the payment system 200 to process payment for the fee corresponding to the applied membership level when registering as a member from the customer terminal 300. The payment processing unit 124 processes, for example, customer authentication, adult authentication, payment method selection, payment request, and payment approval confirmation through member information. Of course, the payment processing unit 124 may process additional costs when purchasing additional wine other than the wine subscription service according to the membership level, changing wine quality, selecting a gift, etc.

The consultation processing unit 126 is equipped with, for example, an artificial intelligence chatbot engine, and processes inquiries about wine in real time by linking with the customer terminal 300.

And the database unit 104 includes wine information 150, member information 152, wine stylist list 154, survey information 156, survey result data 158, feedback survey information 160, and feedback data 162. ), wine recommendation list (164), and wine-related content (166) are stored and managed.

In this embodiment, the wine information 150 includes various information about each of a plurality of wines, such as wine category (type), wine variety, wine producing country, production region, wine attributes, and wine-related content. . Member information 152 includes, for example, personal information of the customer terminal 300, membership level, wine style, etc. Additionally, the rotation information 152 further includes subscription wine, purchase points, wine subscription history, etc. The wine style list 154 includes setting values for each wine to classify wines by matching them to a predefined wine style, for example, a classification value for a wine category, an attribute value for each wine attribute, and a wine category. and weights for wine attributes. The survey information 156 includes a survey of wine category information and wine attribute information, such as the customer's preferred wine type, wine variety, and country of production. The survey result data 158 includes information such as wine category classification values and wine attribute values input from the customer terminal 300 in response to the survey information 156. The feedback survey information 160 includes a survey on the attribute values of the recommended wine, the evaluation and satisfaction of customers who subscribed to the recommended wine, etc. The feedback data 162 includes information input from the customer terminal 300 in response to the feedback survey information 160. The feedback data 162 includes substantially similar or identical content to the survey result data 158. The wine recommendation list 164 includes information about wines that match the customer's wine style. And the wine-related content 166 includes various information about each wine, such as producer, product name, variety, country of origin, vintage, grade, alcohol content, type, drinking temperature, decanting, food, evaluation, and price. , reviews, etc. are included.

Next, the processing procedure of the wine recommendation system according to the present invention will be described using FIGS. 3 to 5.

Figure 3 is a flowchart showing the wine recommendation procedure of the wine recommendation system using wine style matching according to the present invention, and Figure 4 is a flowchart showing the procedure for recommending wine based on the wine questionnaire from the first customer shown in Figure 3. , FIG. 5 is a flowchart showing a procedure for recommending wine based on feedback data about the recommended wine from the customer shown in FIG. 3. These procedures are processed by the components of the wine recommendation system shown in Figures 1 and 2.

3 to 5, in the wine recommendation system 2 using wine style matching of the present invention, in step S400, the wine recommendation server 100 selects various types of wine in advance with at least body, acidity, tannin, and sweetness. It is defined and classified into multiple wine styles with reference values based on the wine attributes included.

In an embodiment of the present invention, wine styles are classified into red wine, white wine, and sparkling wine depending on the type of wine, and the first to third red wine styles for red wine and the first to third red wine styles for white wine depending on the standard values of wine properties. The 3rd white wine style is defined and classified into the 1st to 1st sparkling wine styles in sparkling wine. The wine styles defined in this way are stored and managed as a wine style list in the database unit 104 of the wine recommendation server 100.

In step S410, the customer terminal 300 connects to the wine recommendation server 100 through the communication network 10 and registers as a member. When registered as a member, the wine recommendation server 100 surveys the first customer's wine style. The survey information for this purpose is transmitted to the customer terminal 300.

In step S420, the wine recommendation server 100 receives survey result data from the customer terminal 300, analyzes the survey result data using artificial intelligence, and derives and classifies the customer's wine style through this. The wine recommendation server 100 compares and analyzes the wine styles derived and classified from the survey result data with a wine style list including preset wine styles.

In this embodiment, as shown in FIG. 4, the wine recommendation server 100 creates a preset wine style list using survey result data, wine category classification values of the wine style list, wine attribute values, and respective preset weights. Compare and analyze.

Specifically, in order to analyze the survey result data, the wine recommendation server 100 extracts wine attribute values from the survey result data in step S422 and multiplies the extracted wine attribute values by a preset weight to determine the customer's wine attribute values. Calculate your tastes.

At this time, the wine recommendation server 100 calculates the wine attribute style for each wine using the wine attribute values and weights of body, acidity, sweetness, and tannin, as shown in Equation 1 below.

Here, each of the wine attributes, namely body, acidity, sweetness, and tannin, is set to a 5-level attribute value, n is the number of wines, and X _i is the body, acidity, sweetness, and tannin of the wine attributes among the survey result data. This is the survey value for each, Z _i is the set value for each of the wine attributes set in the wine style list, such as body, acidity, sweetness, and tannin, and the integer 0.2 is the set deviation value between the survey value and the set value. Therefore, the deviation between the customer preference value and the attribute value of each wine according to the survey result data is applied, and the average value of the wine attribute values is calculated to analyze the wine attribute style according to the customer's taste.

In step S424, the wine recommendation server 10 extracts a wine category classification value from the survey result data, and calculates the customer's taste for the wine category classification value by multiplying the extracted wine category classification value by a preset weight.

At this time, the wine recommendation server 100 analyzes the wine category style for each wine using the wine category classification value and preset weight for each wine type, variety, and country, as shown in Equation 2 below.

Here _, n is _the number of wines, At this time, the weight is increased according to the number of wine categories selected by the customer, and weight is given to wines corresponding to the wine category classification value selected in the survey result data. If a wine category is not selected in the survey result data, the weight of the wine attribute value is set to 1 and the weight of the wine category classification value is set to 0.

Therefore, the wine style of each wine is created by adding the product of the wine attribute value and the weight and the product of the wine category classification value and the weight according to Equation 1 and Equation 2.

In step S426, the wine recommendation server 100 analyzes tastes according to wine attribute values and wine category classification values, and classifies a wine style suitable for the customer's tastes through this. Subsequently, in step S428, the classified wine style is compared with a preset wine style list.

Referring again to FIG. 3, in step S430, the wine recommendation server 100 selects at least one wine style preferred by the customer according to the results of comparing and analyzing the wine style derived and classified from the survey result data and the preset wine style list. Create and save a wine recommendation list that includes.

In step S440, the wine recommendation server 100 automatically selects a wine that matches at least one wine style from the wine recommendation list through artificial intelligence and recommends it to the customer terminal 300. At this time, the wine recommendation server 100 provides wine-related content about the recommended wine to the customer terminal 300. Additionally, the wine recommendation server 100 may transmit a wine recommendation list to the customer terminal 300 and provide the customer terminal 300 with a choice of one wine.

Accordingly, when the wine recommendation server 100 subscribes to a wine recommended from the customer terminal 300 or selects a wine, it transmits feedback survey information to the customer terminal 300 and notifies the customer terminal 300 of this. At this time, the wine recommendation server 100 can further consult with the customer terminal 300 about the recommended wine, and when the final wine is decided for subscription, payment can be processed according to the membership level.

In step S450, the wine recommendation server 100 is connected after the customer terminal 300 subscribes to the recommended wine, that is, after the recommended wine is delivered to the customer and the customer experiences the recommended wine, the wine recommendation server When connected to 100, the wine recommendation server 100 provides feedback survey information to the customer terminal 300 in step S460.

In step S470, the wine recommendation server 100 receives feedback data in response to feedback questionnaire information from the customer terminal 300, analyzes the customer's wine style, and reclassifies it. At this time, the reclassified wine style may be the same as the initially classified wine style, or it may change after the wine subscription.

In this embodiment, the wine recommendation server 100 compares and analyzes a preset wine style list using feedback data, wine attribute values of the wine style list, and weights according to time, as shown in FIG. 5.

Specifically, in order to analyze the feedback data, the wine recommendation server 100 extracts wine attribute values from the feedback data in step S472, and multiplies the extracted wine attribute values by a preset weight according to time to determine the wine attribute values. Recalculate customer tastes.

At this time, the wine recommendation server 100 applies the feedback survey value to each wine attribute value of body, acidity, sweetness, and tannin for the recommended wine, as shown in Equation 1, to determine wine attributes according to the customer's taste. Re-analyze the style. Here, the feedback survey value can be added or subtracted compared to the wine attribute value (for example, a range value between -2 and +2 for 5 grades), and the absolute value of the wine attribute value plus the survey value fed back is the customer's taste value. Calculated as In addition, the deviation between the feedback customer preference value and the attribute value of each wine is applied, and the average value of the wine attribute values is calculated.

In step S474, the wine recommendation server 100 reanalyzes the wine attribute style according to the customer's taste by applying weight to the wine attribute value according to the customer's taste over time.

At this time, the wine recommendation server 100 recalculates a wine attribute style suitable for the customer's taste by multiplying the wine attribute value by a weight according to time, as shown in Equation 3. Here, the weight according to time is set higher for the latest feedback data.

Here, the left term is the average value of wine attribute values, and the right term is the weight according to time. The weight is set to 1 when it is the latest feedback data, and takes on values between 1 and 0 as time passes. In this case, if more than 50 days have passed, the weight is applied as 0.

Therefore, the wine style of each wine is reproduced by calculating the product of the wine attribute value and the time-dependent weight according to Equation 1 and Equation 3.

In step S476, the wine recommendation server 100 reanalyzes the taste according to the wine attribute values and reclassifies the wine style suitable for the customer's taste through this. Subsequently, in step S478, the reclassified wine style is re-compared with the preset wine style list.

Referring again to FIG. 3, in step S480, the wine recommendation server 100 receives feedback data from the customer terminal 300, reanalyzes and reclassifies the customer's wine style, and selects the wine style classified from the feedback data as a preset wine. Update the customer's wine recommendation list by comparing and analyzing it with the style list.

Subsequently, in step S490, the wine recommendation server 100 automatically selects a wine from the updated wine recommendation list and recommends it to the customer terminal 300, or transmits the updated wine recommendation list to the customer terminal 300 so that the customer can select one wine. Choose a wine.

Therefore, the wine recommendation server 100 of the present invention uses an artificial intelligence wine recommendation algorithm to derive a wine style suitable for the customer's taste based on a survey, and compares and matches the derived wine style with a predefined wine style. It recommends wines that correspond to at least one wine style, and based on feedback surveys, the wine subscription service is repeated to accurately analyze customers' tastes and recommend wines.

In the above, the configuration and operation of the wine recommendation system using wine style matching according to the present invention has been shown in accordance with the detailed description and drawings, but this is merely an example and is not departing from the technical spirit of the present invention. Various changes and modifications are possible.

2: Wine recommendation system
10: communication network
100: Wine recommendation server
102: server part
104: database section
200: customer terminal

Claims (5)

In a wine recommendation system:
A wine recommendation server that provides a wine subscription service that allows customers registered as members through a communication network to purchase recommended wines by regularly subscribing to wine online.
The wine recommendation server is,
Multiple wines are classified by setting classification values for each wine category and attribute values for multiple wine properties, defining the wine style for each wine as a standard value according to the wine properties, and including the defined wine styles. a wine information management department that creates a list of wine styles;
When registering as a member of the customer terminal, a customer management unit that provides different types of wine subscription services according to membership level;
a payment processing unit that processes payment for a wine subscription service according to the membership level of the customer terminal by linking the customer terminal with an external payment system;
When first registering as a member of the customer terminal, survey information about the wine category and wine attributes is provided to the customer terminal, and survey result data corresponding to the survey information is received from the customer terminal to derive the customer's preferred wine style and A survey processing unit that classifies and generates the derived and classified wine styles as a customer's wine style list;
Provide feedback survey information about the wine recommended to the customer terminal, receive feedback data corresponding to the feedback survey information from the customer terminal, reclassify the customer's wine style, and convert the reclassified wine style into the customer's wine style. A feedback processing unit that updates the list;
Compare the wine style list generated from the survey processing unit with the wine style list preset from the wine information management unit and recommend at least one wine corresponding to a predefined wine style that matches the customer's preferred wine style to the customer terminal. and a wine recommendation processing unit that compares the wine style list updated from the feedback processing unit with a preset wine style list and recommends the next wine to the customer terminal. and
It includes a control unit that controls the wine information management unit, the customer management unit, the payment processing unit, the survey processing unit, the feedback processing unit, and the wine recommendation processing unit to process the operation of the wine recommendation server in conjunction with the customer terminal and the payment system. A wine recommendation system characterized by:
In claim 1,
The survey processing department,
Wine attribute values are extracted from survey result data, preset weights are applied to the extracted wine attribute values to calculate survey values for the customer's wine attributes, and the wine attribute survey values and wine attribute values from the preset wine style list are calculated. Apply the deviation from and calculate the average of the wine attribute survey values to analyze the wine attribute style according to the customer's wine attribute taste;
Wine category classification values are extracted from the survey result data, and a preset weight is applied to the extracted wine category classification values to analyze the wine category style according to the customer's wine category taste; next
A wine recommendation system that classifies and creates a customer's wine style list through the analyzed wine attribute style and wine category style.
In claim 1,
The feedback processing unit,
Wine attribute values are extracted from feedback data, preset weights are applied to the extracted wine attribute values to recalculate customer wine attribute survey values, and wine attribute survey values and wine attribute values from the preset wine style list are calculated. It is characterized by calculating the average of the survey values of wine attributes by applying the deviation from and reanalyzing the customer's wine attribute style by applying weights according to time to the survey values of wine attributes and updating the customer's wine style list. Wine recommendation system.
In claim 3,
The wine recommendation processing unit,
Using artificial intelligence, survey result data or feedback data are analyzed to create or update a wine recommendation list corresponding to the customer's wine style list;
A wine recommendation system that automatically selects and recommends one wine from a created or updated wine recommendation list and transmits it to the customer terminal, or transmits a wine recommendation list to the customer terminal to select a wine.
The method according to any one of claims 1 to 4,
The wine information management department,
Depending on the various types of wine, they are classified into red wine, white wine, and sparkling wine, and according to the standard values of wine properties that include at least the body, acidity, tannin, and sweetness of the wine, each of which is classified into 5 grades, red wine is classified into 1st to 5th grade wines. The 3rd red wine style is defined and classified as the 1st to 3rd white wine style for white wine, and the 1st to 3rd sparkling wine style for sparkling wine;
The first red wine style is defined as wines with standard values of wine properties such as body grade 4 or higher, acidity grade 4 or higher, tannin grade 4 or higher, and sweetness ignored, and at the same time, body, acidity, and tannin. This also applies to wines where one of the reference values is 3 stars but the other two are 5 stars;
The second red wine style is defined as wines that have standard values of wine properties such as body grade 3 or higher, acidity grade 3 or higher, tannin grade 3 or higher, and sweetness ignored, and at the same time, body, acidity, and tannin. It does not apply to wines in which any one of the standard values is below grade 2, excludes wines in which both body and tannin are grade 3, and applies to all wines excluding the first red wine style among wines that satisfy all of the above conditions. become;
The third red wine style is defined as wines that have body of 3 or less grades, acidity of 4 or more, tannin of 3 or less, and sweetness ignored as standard values for wine properties, and at the same time, body, acidity and tannin are defined as standard values. Applies to wines that meet all criteria;
The first white wine style is defined as wines having a body of 3 or more grades, acidity of 4 or more, and tannin and sweetness ignored as standard values for wine properties;
The second white wine style is defined as wines with standard values of wine properties: body of 3 or higher, acidity of 3 or lower, and tannin and sweetness ignored;
The third white wine style is defined as wines with standard values of wine properties: body of 2 grades or less, acidity of 4 grades or more, and tannin and sweetness ignored;
The first sparkling wine style is defined as wines that have a sweetness level of 3 or higher as a standard value of wine attributes and the standard values of other wine attributes are ignored; and
The second sparkling wine style is a wine recommendation system characterized in that it is defined as wines with a sweetness level of 2 or less as a standard value of wine attributes and the standard values of other wine attributes are ignored.