KR102452667B1 - Apparatus of alcoholic liquors recommendation using purchase history - Google Patents

Abstract

The present invention relates to an alcoholic beverage recommendation device using a purchase history, and the alcoholic beverage recommendation device using a purchase history according to an aspect of the present invention generates a user identification code that generates a user identification code to distinguish a user from a Thai user according to a user input. We collect purchase history information for each user, including the purchase product name, purchase quantity, and purchase date and time, from a preset judgment point to a point in time prior to a certain period of time as a purchase history following a purchase by a user selecting alcoholic beverages. Purchase history information collection unit that matches user identification code, alcoholic beverage product information DB that stores alcoholic beverage product information including alcohol product name, alcohol type, alcohol content, and flavor for a plurality of alcoholic beverage products, The total purchase frequency value is calculated by summing the purchase quantity of a plurality of purchase history information, and corresponds to a plurality of purchase history information among a plurality of alcoholic beverage characteristics including a plurality of preset alcoholic beverage types, frequency ranges, and flavors, respectively. The purchase frequency value for each alcoholic beverage characteristic is calculated by counting the number of purchases included in each of a plurality of purchase history information for each alcoholic beverage characteristic corresponding to any one of the type, alcohol content, and flavor of the alcoholic beverage product information, and the total purchase frequency A purchase ratio calculation module for calculating the purchase ratio for each alcoholic beverage characteristic according to the ratio of the purchase frequency value for each alcoholic beverage characteristic to the value; set as a weight value, and a value obtained by multiplying an initial value by a predetermined value corresponding to the purchase ratio for an alcoholic beverage characteristic in which a purchase ratio is equal to or greater than a predetermined threshold value among the plurality of alcoholic beverage characteristics is set as a weight value; A preference information generating unit that generates preference information including a weight value for each characteristic and matches the user identification code with the user identification code; The interface is provided to the user, and the keywords input through the user interface are transmitted as a query to the mainstream product information DB, and the A search unit that receives a search result including a plurality of alcoholic product information from the alcoholic beverage product information DB as a response, and an alcoholic beverage product that provides a list of alcoholic beverages by listing a plurality of alcoholic product information included in the search result based on preference information Includes a list providing module.

Description

Alcohol recommendation device using purchase history {APPARATUS OF ALCOHOLIC LIQUORS RECOMMENDATION USING PURCHASE HISTORY}

The present invention relates to an alcoholic beverage recommendation apparatus using a purchase history, and more particularly, to an alcoholic beverage recommendation apparatus using a purchase history capable of recommending an alcoholic beverage preferred by a user using the user's alcoholic beverage purchase history.

Alcohol consumption culture is changing from consuming a large amount of a small number of products centered on a conventional group meeting to consuming a small amount of various products by individuals or small groups.

As a result, alcoholic beverage products with various flavors and strengths are being developed to satisfy consumer preferences, but due to the limitation of information on alcoholic beverages, users must directly taste various alcoholic beverages to determine whether the product suits their preference. There is a problem in that it is difficult to find a liquor product that meets individual preferences.

In particular, there is a problem in that alcohol products with low awareness or produced in small quantities are only available in some regions or stores, making it difficult for users to even access the products.

Republic of Korea Patent Application No. 10-2020-0091220 Republic of Korea Patent Application No. 10-2018-0003321

It is an object of the present invention to solve the above problems, by identifying each user's preference for alcoholic beverage characteristics such as the type of alcoholic beverage, alcohol content, and flavor from the user's actual purchase history, and ordering alcoholic beverage products in order of preference by the user It is to provide an alcoholic beverage recommendation device using a purchase history that can provide information.

Objects of the present invention are not limited to the objects mentioned above, and other objects not mentioned will be clearly understood from the description below.

A liquor recommendation apparatus using a purchase history according to an aspect of the present invention for achieving the above object includes a user identification code generating unit that generates a user identification code for distinguishing a user from a Thai user according to a user's input, and a user drinking alcohol. Purchasing history information including the purchase product name, purchase quantity, and date of purchase made from a preset judgment point to a point before a certain period as a purchase history according to selection and purchase is collected for each user and matched with the user identification code Purchasing history information collection unit, alcoholic beverage product information DB for storing alcoholic beverage product information including alcohol product name, alcoholic beverage type, alcohol content, and flavor for a plurality of alcoholic beverage products, and a plurality of purchase history information matching user identification codes. The total purchase frequency value is calculated by adding up the purchase quantity, and alcoholic beverage product information corresponding to a plurality of purchase history information among a plurality of alcoholic beverage characteristics including a plurality of preset types of alcoholic beverages, frequency ranges, and flavors. A purchase frequency value for each alcoholic beverage characteristic is calculated by counting the number of purchases included in each of a plurality of purchase history information for each alcoholic beverage characteristic corresponding to any one of type, alcohol content, and flavor, and the alcoholic beverage characteristic with respect to the total purchase frequency value A purchase ratio calculation module that calculates the purchase ratio for each alcoholic beverage characteristic according to the ratio of each purchase frequency value, and sets a preset initial value as a weight value for the alcoholic beverage characteristic in which the purchase ratio among the plurality of alcoholic beverage characteristics is less than a predetermined threshold value; A value obtained by multiplying an initial value by a predetermined value corresponding to the purchase ratio for an alcoholic beverage characteristic with a purchase ratio equal to or greater than a predetermined threshold value among the plurality of alcoholic beverage characteristics is set as a weight value, and a weight value for each of the plurality of predetermined alcoholic beverage characteristics A preference information generating unit that generates preference information including and matches the user identification code, and provides a user with a predetermined user interface to receive a keyword including at least one of product name, type of alcoholic beverage, alcohol content, and flavor from the user, and Keywords input through the user interface are transferred to the mainstream product information DB as a query, and as a response to the query, the mainstream product information DB and a search unit for receiving a search result including a plurality of alcoholic product information from .

According to the present invention, it is possible to identify each user's preference for alcoholic beverage characteristics such as alcoholic beverage type, alcohol content, and flavor from the user's actual purchase history, and obtain a purchase history that can provide information on alcoholic beverages in the order that the user is most likely to prefer. There is an effect of providing a used alcoholic beverage recommendation device.

According to the present invention, by using the preference of other users for the characteristics of the alcoholic beverage, it is possible to recommend alcoholic beverage products corresponding to the characteristics of the alcoholic beverage that the user prefers even though the purchase history is relatively small.

In addition, according to the present invention, by using location information in the recommendation process to provide additional information, such as information on a liquor store in which a specific product is located or information on a nearby liquor store, it is possible to increase the user's convenience in purchasing.

The effects of the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those skilled in the art from the description of the claims.

1 is a block diagram of an apparatus for recommending alcoholic beverages using a purchase history according to an embodiment of the present invention.
Figure 2a is a graph showing the correlation between alcoholic beverage characteristics beer and alcoholic beverage characteristics soju according to embodiments of the present invention.
Figure 2b is a graph showing the correlation between alcoholic beverage characteristics beer and alcoholic beverage characteristics according to embodiments of the present invention.
Figure 2c is a graph showing the correlation between alcoholic beverage characteristics beer and alcoholic beverage characteristics hop flavor according to embodiments of the present invention.
3 is a flowchart of a method for recommending alcoholic beverages using a purchase history according to another embodiment of the present invention.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but will be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and those of ordinary skill in the art to which the present invention pertains. It is provided to fully inform the person of the scope of the invention, and the present invention is only defined by the description of the claims. On the other hand, the terms used in the present specification are for describing the embodiments, and are not intended to limit the present invention. In this specification, the singular also includes the plural, unless specifically stated otherwise in the phrase.

1 is a block diagram of an apparatus for recommending alcohol using a purchase history according to an embodiment of the present invention. Referring to FIG. 1 , the apparatus for recommending alcohol using a purchase history according to an embodiment of the present invention 10 generates a user identification code. Unit 101, purchase history information collection unit 103, alcoholic beverage product information DB 105, purchase ratio calculation module 110, preference information generation unit 120, search unit 130, alcoholic beverage product list providing module ( 140 ), a correlation analysis unit 150 , an alcoholic beverage product selection unit 160 , a location information collection unit 170 , and an alcoholic beverage store information acquisition unit 180 .

The user identification code generating unit 101 may generate a user identification code for distinguishing a user from a tie user according to a user's input.

The user identification code generation unit 101 provides an interface for receiving user information including at least one of the user's name, gender, date of birth, ID, and password, and selects the user based on the user information input according to the user's input. It may be to generate a user identification code for distinguishing from the tie user.

The purchase history information collection unit 103 is a purchase history according to which a purchase occurs by selecting alcoholic beverages by the user, from the predetermined time of judgment to the time before a certain period (for example, 2 months), the purchase product name, purchase quantity, Purchasing history information including at least one of a purchase date and time, purchase location, and purchase price may be collected for each user and matched with a user identification code.

Here, the preset judgment time may be any one time point after the user first purchases the alcoholic beverage product.

The determination time may be reset every preset period (eg, one month) from an initial time set according to a user's input.

For example, if the judgment point is set to May 1, 2020 according to the user's input and the preset period is one month, the judgment point after May 1, 2020 is reset to June 1, 2020, and the The judgment point after January 1 may be reset to July 1, 2020.

The purchase history information collection unit 103 collects purchase history information for each user, which is between the point of time when the purchase date and time is determined from the time of determination based on a predetermined time of determination, and a point before a predetermined period (for example, 3 months) from the time of determination, and identifies the user. It may be to match the code.

For example, if the preset period is 3 months and the purchase history information is collected at 18:00 on September 10, 2020, the purchase history information collection unit 103 starts from 18:00 on June 11, 2020 It may be to collect purchase history information including purchase date and time between September 2020, 18:00 for each user and match it with a user identification code.

Table 1 below is an example of a plurality of purchase history information matched for any one user identification code.

Date of purchase Purchase product name purchase quantity purchase price per piece 20/06/09 16:00 ○○○ 2 4100 20/06/12 18:23 △ △ △ 2 3500 20/07/08 18:30 ○○○ 3 4100 20/07/11 18:44 ☆☆☆ 2 4300 20/07/13 19:12 □□□ One 4800 20/07/28 19:05 ◇◇◇ 2 2300 20/08/04 18:53 ◇◇◇ 3 2300 20/08/16 20:08 ○○○ 2 4100 20/08/30 19:53 ◇◇◇ 2 2300 20/09/01 19:49 ◇◇◇ One 2300

The purchase history information collection unit 103 collects purchase history information on which purchases have been made recently, and reflects the user's changing preferences over time.

The alcoholic beverage product information DB 105 may store alcoholic beverage product information including the alcohol product name, alcoholic beverage type, alcohol content, and flavor for a plurality of alcoholic beverage products.

Table 2 below is prepared as an example to describe a plurality of alcoholic product information stored in the alcoholic beverage product information DB 105 , but is not limited thereto.

Liquor product name alcoholic beverages Frequency (%) flavor ○○○ beer 7 H, C △ △ △ soju 15 F ☆☆☆ fruit wine 20 F □□□ beer 4 H, C, F ◇◇◇ sparkling wine 3 C, F

Flavor H in Table 2 means hop flavor, C means carbonic acid, and F means fruit flavor.

The purchase ratio calculation module 110 calculates a total purchase frequency value by summing the purchase quantities of a plurality of purchase history information matched to each user identification code, and includes a plurality of preset alcoholic beverage types, frequency ranges, and flavors Counting the number of purchases included in each of the plurality of purchase history information for each alcoholic beverage characteristic corresponding to any one of the alcoholic beverage type, alcohol content, and flavor of the alcoholic beverage product information corresponding to the plurality of purchase history information among the plurality of alcoholic beverage characteristics Accordingly, the purchase frequency value for each alcoholic beverage characteristic may be calculated, and the purchase ratio for each alcoholic beverage characteristic may be calculated according to the ratio of the purchase frequency value for each alcoholic beverage characteristic to the total purchase frequency value.

Here, the alcoholic beverage product information corresponding to each purchase history information means alcoholic beverage product information including the purchased product name and the corresponding alcoholic beverage product name in the purchase history information.

The purchase ratio calculation module 110 may include a first purchase frequency calculation unit 111 , a second purchase frequency calculation unit 113 , and a purchase ratio calculation unit 115 .

The first purchase frequency calculation unit 111 may calculate the total purchase frequency value by summing the purchase quantities of a plurality of purchase history information matched to each user identification code.

For example, if a plurality of purchase history information matched to each user identification code is the same as in Table 1, the first purchase frequency calculation unit 111 calculates a total purchase frequency value of 20 by summing the purchase quantities of the plurality of purchase history information. may be doing

Table 3 below shows the purchase date and quantity of purchase history information in Table 1 above for convenience of explanation and the type, alcohol content, and flavor of alcoholic beverage product information corresponding to each purchase history information among alcoholic beverage product information in Table 2 it will be listed

Date of purchase Purchase (Alcoholic) Product Name purchase quantity alcoholic beverages Frequency (%) flavor 20/06/14 16:00 ○○○ 2 beer 7 H, C 20/06/21 18:23 △ △ △ 2 soju 15 F 20/07/08 18:30 ○○○ 3 beer 7 H, C 20/07/11 18:44 ☆☆☆ 2 fruit wine 20 F 20/07/13 19:12 □□□ One beer 4 H, C, F 20/07/28 19:05 ◇◇◇ 2 sparkling wine 3 C, F 20/08/04 18:53 ◇◇◇ 3 sparkling wine 3 C, F 20/08/16 20:08 ○○○ 2 beer 7 H, C 20/08/30 19:53 ◇◇◇ 2 sparkling wine 3 C, F 20/09/01 19:49 ◇◇◇ One sparkling wine 3 C, F

The second purchase frequency calculation unit 113 is configured to determine the type of alcoholic beverage, the alcohol content, For each alcoholic beverage characteristic corresponding to any one of the flavors, the purchase frequency value for each alcoholic beverage characteristic may be calculated by counting the number of purchases included in each of a plurality of purchase history information.

For example, referring to Table 3, the second purchase frequency calculation unit 113 may include '○○○' and '□□□' and '○○○' and '□□□' for the alcoholic beverage product information corresponding to the plurality of purchase history information of Table 1, for which the type of alcoholic beverage is beer. By counting the purchase quantity of ', it may be to calculate a purchase frequency value of 8 for beer characteristic of alcoholic beverages.

Table 4 below shows alcoholic beverages calculated for a plurality of alcoholic beverage characteristics including a plurality of alcoholic beverage types, frequency ranges, and flavors preset by the second purchase frequency calculation unit 113 for a plurality of purchase history information of Table 1 The purchase frequency value for each characteristic is shown.

alcoholic beverage Purchasing Frequency Value by Alcohol Characteristics alcoholic beverages beer 8 soju 2 fruit wine 2 liquor 0 sparkling wine 8 Frequency range (%) 0 or more and less than 5 9 5 or more and less than 10 7 over 10 4 flavor H 6 C 16 F 11

The characteristics of alcoholic beverages in Table 4 are for convenience of explanation, and are not limited thereto.

The purchase ratio calculation unit 115 is a liquor characteristic according to the ratio of the purchase frequency value for each alcoholic beverage characteristic calculated by the second purchase frequency calculation unit 113 to the total purchase frequency value calculated by the first purchase frequency calculation unit 111 . It may be to calculate a star purchase ratio.

The purchase ratio calculation unit 115 may calculate the purchase ratio for each alcoholic beverage characteristic by multiplying a value obtained by dividing the purchase frequency value for each alcoholic beverage characteristic by the total purchase frequency value by 100.

For example, referring to Table 4, when the total purchase frequency value is 20, the purchase ratio calculation unit 115 calculates a purchase ratio of 40% for alcoholic beverages characteristic beer, and calculates a purchase ratio of 10% for alcoholic beverage characteristic soju. , calculating a purchase ratio of 10% for fruit wine with alcoholic beverage characteristics, calculating a purchase ratio of 0% for alcoholic beverage characteristic spirits, and calculating a purchase ratio of 40% for spirited alcoholic beverages.

The purchase ratio calculated by the purchase ratio calculation unit 115 for each alcoholic beverage characteristic of Table 4 is summarized in Table 5 below.

alcoholic beverage Purchasing Frequency Value by Alcohol Characteristics Purchase rate (%) alcoholic beverages beer 8 40 soju 2 10 fruit wine 2 10 liquor 0 0 sparkling wine 8 40 Frequency range (%) 0 or more and less than 5 9 45 5 or more and less than 10 7 35 over 10 4 20 flavor H 6 30 C 16 80 F 11 55

In the case of flavor, since a plurality of flavors may be included in one alcoholic beverage product, the total sum of purchase ratios of alcoholic beverage characteristics corresponding to flavor may exceed 100%.

The preference information generating unit 120 sets a preset initial value (for example, 1) as a weight value for the alcoholic beverage characteristic in which the purchase ratio among the plurality of alcoholic beverage characteristics is less than a predetermined threshold value (eg 20%), A value obtained by multiplying an initial value by a predetermined value corresponding to the purchase ratio for an alcoholic beverage characteristic in which a purchase ratio is greater than or equal to a predetermined threshold value (eg, 20%) among a plurality of alcoholic beverage characteristics is set as a weight value, It may be to generate preference information including a weight value for each mainstream characteristic and match it with the user identification code.

Here, the preset value corresponding to the purchase ratio is a value proportional to the purchase ratio, and may be a value predetermined for each purchase ratio range according to the size of the purchase ratio, or a value obtained by multiplying the purchase ratio by a preset multiple, but is not limited thereto .

For example, when the preset threshold value is 20% and the preset initial value is 1, the preference information generating unit 120 sets a preset initial value for alcohol-characteristic shochu and fruit wine in which the purchase ratio is 10% and less than the preset threshold value. The value 1 may be set as a weight value, and a value obtained by multiplying an initial value by a value proportional to the purchase ratio for alcoholic beer and sparkling wine having a purchase ratio of 40% or more, which is a preset threshold value or more, may be set as a weight value.

The preference information generating unit 120 may generate the user's preference information at each determination point based on the purchase history information for each user obtained based on each determination point reset at each predetermined period.

The preference information generating unit 120 calculates the total calculated by the first purchase frequency calculation unit 111 with respect to the purchase history information collected for each user based on the first determination time, which is any one determination time according to a preset period after the initial time point. When the purchase frequency value is greater than or equal to a preset threshold (eg, 50), user preference information at the first determination point may be generated based on purchase history information collected for each user.

The total purchase frequency value calculated by the first purchase frequency calculation unit 111 for purchase history information collected for each user based on the first determination time, which is one determination time according to a predetermined period after the initial time point, is a preset threshold value (for example, less than 50), the purchase history information collection unit 103 when the total purchase frequency value calculated by the first purchase frequency calculation unit 111 for the purchase history information collected for each user exceeds the threshold value Until the purchase history information prior to the period based on the first determination time is further collected, the preference information generation unit 120 is collected for each user until the total purchase frequency value exceeds the threshold value before the first determination time. It may be to generate user preference information at the first determination point based on the purchase history information.

That is, the preference information generating unit 120 determines that the total purchase quantity of purchase history information collected for each user is greater than or equal to a predetermined threshold value based on a determination point reset according to a preset period, based on the purchase history information collected for each user. As the preference information is generated, the user preference information may be updated.

In addition, if the total purchase quantity of purchase history information collected for each user is less than the preset threshold based on the judgment time reset according to the preset period, user preference information is further collected based on the purchase history information until the threshold value is exceeded. It may be to update user preference information according to the creation of .

Accordingly, for users with a short product purchase cycle, preference information is continuously updated based on a relatively recent purchase history, and for users with a long product purchase cycle or with little or no recent purchase history, even a relatively past purchase history is included. The generated preference information may be provided.

The search unit 130 receives at least one of product name, alcoholic beverage type, selling price, alcohol content, and flavor as a keyword from the user, transmits the keyword as a query to the alcoholic beverage product information DB 105, and responds to the query in the alcoholic beverage product information DB It may be to receive a search result including information on a plurality of alcoholic beverages from 105 .

The search unit 130 provides a predetermined search interface to receive at least one of product name, alcoholic beverage type, selling price, alcohol content, and flavor as a keyword from the user, and provides a product name, alcoholic beverage type, and minimum selling price from the user through the search interface. , at least one of a maximum selling price, a minimum frequency, a maximum frequency, and a flavor may be input as a keyword.

For example, the search unit 130 receives the minimum frequency '1' and the maximum frequency '5' as keywords, respectively, and queries the keyword minimum frequency '1' and the maximum frequency '5' to the alcoholic product information DB 105 as a query. and receiving a search result including a plurality of alcoholic product information having a frequency between 1 and 5 from the alcoholic product information DB 105 as a response to the query.

In addition, when the search unit 130 receives the keyword 'shochu' corresponding to the type of alcoholic beverage from the user, the search unit 130 transmits the keyword 'soju' as a query to the alcoholic beverage product information DB 105, and provides information on alcoholic beverages as a response to the query. It may be that a search result including information on a plurality of alcoholic beverages including 'soju' as a type of alcoholic beverage is received from the DB 105 .

The alcoholic beverage product list providing module 140 may provide the alcoholic beverage product list by listing a plurality of alcoholic product information included in the search result input from the search unit 130 based on the preference information.

It may include a preference score calculation unit 141 and a liquor product list generation unit 143 .

In another example, the alcoholic beverage product list providing module 140 receives an alcoholic beverage recommendation request signal for receiving a beverage product recommendation from the user, and stores a plurality of alcoholic beverage product information stored in the alcoholic beverage product information DB 105 based on the user's preference information. alcohol with high user preference by calculating a preference score for It may be a product recommendation.

The preference score calculation unit 141 may calculate a preference score for each alcoholic beverage product included in the search result based on the preference information.

The preference score calculation unit 141 adds a weight value corresponding to the alcohol type and the weight value corresponding to the alcohol in the preference information corresponding to each alcoholic beverage type, alcohol content, and flavor of each alcoholic beverage product included in the search result for each alcoholic beverage product. It may be to calculate a preference score.

The alcoholic beverage list generator 143 may generate the alcoholic beverage product list by listing a plurality of alcoholic product information included in the search result in the order of preference score.

As the alcoholic beverage list generator 143 generates the alcoholic beverage list and outputs it to a predetermined user interface, the alcoholic beverage list generating unit 143 may provide the user with the alcoholic beverage product list, in which the alcoholic beverage product having a high likelihood of user preference is exposed.

In the alcoholic beverage recommendation apparatus 10 using the purchase history according to the embodiment of the present invention, the alcoholic beverage product list providing module 140 calculates a preference score for each alcoholic beverage product included in the search result based on the user's preference information. and create a list of alcoholic beverages by arranging them in ascending order of preference score, so that information on alcoholic beverages with high user preference is exposed at the top.

The correlation analysis unit 150 stores a plurality of preference information matched to a plurality of different user identification codes, respectively, and performs correlation analysis between a plurality of preset mainstream characteristics based on the plurality of preference information. For each of the mainstream characteristics of dogs, a correlation value with other mainstream characteristics may be calculated.

The correlation analysis unit 150 selects a coordinate axis corresponding to a weight value for a first alcoholic beverage characteristic, which is one of the plurality of mainstream characteristics, and a coordinate axis corresponding to a weight value for a second alcoholic beverage characteristic, which is another one of the plurality of alcoholic beverage characteristics. A coordinate region including , 1021) on the coordinate region, and by analyzing the correlation between the first mainstream characteristic and the second mainstream characteristic using a plurality of coordinates 1001, 1011, 1021 on the coordinate region, the first mainstream characteristic and the second mainstream characteristic It may be to calculate a correlation value between mainstream characteristics.

2A is a graph showing the correlation between alcoholic beverage characteristics beer and alcoholic beverage characteristics shochu according to embodiments of the present invention, and FIG. 2B is a graph showing the correlation between alcoholic beverage characteristics beer and alcoholic beverage characteristic spirits according to embodiments of the present invention. 2c is a graph showing the correlation between alcoholic beverage characteristics beer and alcoholic beverage characteristics hop flavor according to embodiments of the present invention, where p is a correlation value between alcoholic beverage characteristics shown in each graph.

At this time, the preference information generating unit 120 may include any one alcoholic beverage characteristic in which the purchase ratio is greater than or equal to a predetermined threshold value (eg, 20%) among the alcoholic beverage characteristics in which the purchase ratio is less than a preset threshold value (eg 20%) and A value obtained by multiplying an initial value by a preset additional weight value may be set as a weight value for a mainstream characteristic in which the correlation value of is greater than or equal to a preset limit value (eg, 0.05).

The preference information generating unit 120 adds an additional weight value to the alcoholic beverage characteristic that has a high correlation with the alcoholic beverage characteristic that the user has purchased relatively frequently among the alcoholic beverage characteristics that the user has little or no purchase history, based on the preference information of other users. You can set the assigned weight value.

For example, the preference information generating unit 120 determines that the purchase ratio is greater than or equal to a predetermined threshold value (eg, 20%) and the correlation value p with beer is a predetermined threshold value (eg, 0.05 or greater, alcohol characteristic hop flavor and Among alcoholic beverages, a value obtained by multiplying an initial value by a predetermined additional weight value may be set as a weight value for a beverage having a purchase ratio of 0% and less than a preset threshold value (eg, 20%).

Accordingly, an additional weight value can be given to alcoholic beverage characteristics that users have a high probability of preferring even though there is little or no purchase history.

The alcoholic beverage selection unit 160 may select any one alcoholic product information included in the alcoholic beverage list according to a user's input.

The location information collecting unit 170 may collect the user's location information from the outside.

The location information collection unit 170 may receive location information corresponding to a point selected by the user from the user, or collect location information from a terminal that is registered in advance in response to a user identification code.

The liquor store information acquisition unit 180 is located within a preset radius based on the user location information from the external DB 2 in which liquor store information including store locations for each of a plurality of liquor stores and the amount of inventory for each liquor product is stored. It may be to obtain and output information about a liquor store where an inventory amount of a liquor product corresponding to the liquor product information selected by the product selection unit 170 is equal to or greater than a preset threshold (eg, 1).

In the alcoholic beverage recommendation apparatus 10 using a purchase history according to an embodiment of the present invention, the alcoholic beverage store information acquisition unit 180 is located near the user or near the point selected by the user, and the stock of the alcoholic beverage product selected by the user is There is an advantage in that by outputting information about a liquor store that exists, it informs a nearby liquor store where the user can purchase the desired product.

In another embodiment, the search unit 130 receives a plurality of preset administrative districts from the user, or receives the user's location information collected by the location information collection unit 170, and receives the input administrative district or the user's location It may be to output information on liquor stores located within a preset radius based on the information, and provide a location search interface for selecting any one of liquor store information from among the information about liquor sellers output according to a user's input.

At this time, the alcoholic beverage product list providing module 140 lists the alcoholic beverage product list in order of preference score for each alcoholic beverage product, which is included in the information on the alcoholic beverage store selected in the location search interface, based on the user's preference information. may be provided.

3 is a flowchart of a method for recommending alcoholic beverages using a purchase history according to another embodiment of the present invention.

The alcoholic beverage recommendation method using the purchase history according to another embodiment of the present invention may be performed by the alcoholic beverage recommendation apparatus 10 using the purchase history according to the embodiment of the present invention.

Hereinafter, the content and configuration overlapping those of the alcoholic beverage recommendation device 10 using the purchase history described above correspond to reference numerals, and detailed descriptions will be omitted for convenience of description.

The alcoholic beverage recommendation device 10 using the purchase history generates a user identification code to distinguish the user from the Thai user according to the user's input (S101), and sets a purchase history as the purchase history when the user selects alcoholic beverages and makes a purchase. A plurality of purchase history information for each user is collected by collecting purchase history information including the purchase product name, purchase quantity, and purchase date and time for each user from the time of judgment to a point prior to a certain period of time and matching it with the user identification code (S103) .

Thereafter, the alcoholic beverage recommendation apparatus 10 using the purchase history calculates the total purchase frequency value by summing the purchase quantities of a plurality of purchase history information matched to the user identification code, and the alcohol product name for a plurality of pre-stored alcoholic beverage products , on the basis of a plurality of purchase history information matched with a user identification code and alcoholic product information corresponding to a plurality of purchase history information among alcoholic product information including alcoholic beverage types, alcohol content, and flavor as contents, a plurality of predetermined alcoholic beverage types, frequency ranges, The purchase included in each of the plurality of purchase history information for each alcoholic beverage characteristic corresponding to any one of the alcoholic beverage type, the alcohol content, and the flavor of the alcoholic beverage product information corresponding to the plurality of purchase history information among the plurality of alcoholic beverage characteristics including flavors As the quantity is counted, the purchase frequency value for each alcoholic beverage characteristic is calculated, and the purchase ratio for each alcoholic beverage characteristic is calculated according to the ratio of the purchase frequency value for each alcoholic beverage characteristic to the total purchase frequency value (S105).

The alcoholic beverage recommendation device 10 using the purchase history sets a preset initial value as a weight value for the alcoholic beverage characteristic in which the purchase ratio is less than a predetermined threshold value among the plurality of alcoholic beverage characteristics, and the purchase ratio among the plurality of alcoholic beverage characteristics is A value obtained by multiplying an initial value by a predetermined value corresponding to the purchase ratio for alcoholic beverage characteristics that is greater than or equal to a set threshold value is set as a weight value, and preference information including weight values for each of a plurality of predetermined alcoholic beverage characteristics is generated to identify users By matching the code, preference information for each user is generated (S107).

Here, the preset value corresponding to the purchase ratio may be a value proportional to the purchase ratio. That is, as the purchase ratio increases, a larger value may be set as a value corresponding to the purchase ratio.

The alcoholic beverage recommendation device 10 using the purchase history provides a user with a predetermined user interface to receive a keyword including at least one of product name, alcoholic beverage type, selling price, alcohol content, and flavor from the user, and inputs the keyword through the user interface. Receive (S109).

Thereafter, the alcoholic beverage recommendation apparatus 10 using the purchase history searches for a plurality of alcoholic product information pre-stored as a query using the input keyword, and inputs a search result including a plurality of alcoholic product information as a response to the query. and provides a list of alcoholic beverages by listing a plurality of information on alcoholic beverages included in the search result based on preference information (S111).

According to embodiments of the present invention, the preference for a plurality of alcoholic beverage characteristics such as the type of alcoholic beverage, the alcohol content, and the flavor may be quantified based on the user's actual purchase history.

In addition, when a user performs a search for alcoholic beverages, it is possible to provide a list of alcoholic beverages in which information on alcoholic beverages that users are likely to prefer is exposed at the top by using the numerical preference for alcoholic beverage characteristics.

According to the embodiments of the present invention, there is an advantage that the user can access various products suitable for his/her preference by being provided with a list of alcoholic beverages reflecting his/her preference based on actual data such as a purchase history.

In addition, preference information for a plurality of users according to embodiments of the present invention can be utilized as data for new product development and marketing.

Those of ordinary skill in the art to which the present invention pertains will understand that the present invention may be embodied in other specific forms without changing the essential features of the technical idea. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the claims to be described later rather than the above detailed description, and all changes or modifications derived from the claims and their equivalent concepts should be construed as being included in the scope of the present invention.

10: Alcohol recommendation device using purchase history
101: user identification code generation unit
103: purchase history information collection unit
105: Alcohol product information DB
110: purchase ratio calculation module
120: preference information generating unit
130: search unit
140: module for providing list of alcoholic beverages
150: correlation analysis unit
160: liquor product selection unit
170: location information collection unit
180: liquor store information acquisition unit

Claims (4)

a user identification code generation unit for generating a user identification code for distinguishing a user from a tie user according to a user's input;
As a purchase history following a purchase by a user selecting alcoholic beverages, purchase history information including the name of the purchased product, the quantity purchased, and the purchase date and time from the predetermined time of judgment to the time before a certain period is collected for each user to identify the user a purchase history information collecting unit that matches the code;
Alcohol product information DB for storing alcoholic beverage product information including alcohol product name, alcoholic beverage type, alcohol content, and flavor for a plurality of alcoholic beverage products;
The total purchase frequency value is calculated by summing the purchase quantities of a plurality of purchase history information matched to the user identification code, and among a plurality of alcoholic beverage characteristics including a plurality of preset types of alcoholic beverages, frequency ranges, and flavors, the Purchasing by alcoholic beverage characteristics by counting the number of purchases included in each of the plurality of purchase history information for each alcoholic beverage characteristic corresponding to any one of the type, intensity, and flavor of the alcoholic beverage product information respectively corresponding to the plurality of purchase history information a purchase ratio calculation module for calculating a frequency value and calculating a purchase ratio for each alcoholic beverage characteristic according to a ratio of the purchase frequency value for each alcoholic beverage characteristic to the total purchase frequency value;
A predetermined initial value is set as a weight value for the alcoholic beverage characteristic in which the purchase ratio is less than a predetermined threshold value among the plurality of alcoholic beverage characteristics, and the purchase ratio among the plurality of alcoholic beverage characteristics is greater than or equal to a predetermined threshold value. for a value obtained by multiplying the initial value by a predetermined value in response to the purchase ratio for a weight value, and generating preference information including a weight value for each of a plurality of predetermined alcoholic characteristics to match the user identification code preference information generator;
A predetermined user interface is provided to the user to receive a keyword including at least one of product name, alcoholic beverage type, alcohol content, and flavor from the user, and the keyword input through the user interface is transmitted as a query to the alcoholic beverage product information DB, a search unit receiving a search result including a plurality of alcoholic product information from the alcoholic beverage product information DB as a response to the query;
a liquor product list providing module for providing a list of alcoholic beverages by listing a plurality of alcoholic product information included in the search result based on the preference information; and
Storing a plurality of preference information matched to a plurality of different user identification codes, performing correlation analysis between a plurality of preset mainstream characteristics based on the plurality of preference information, and providing each of the plurality of mainstream characteristics Correlation analysis unit that calculates a correlation value with other mainstream characteristics for
The preference information generating unit
An additional weight value preset to the initial value for an alcoholic beverage characteristic in which the purchase ratio is equal to or greater than a predetermined threshold value among alcoholic beverage characteristics whose purchase ratio is less than a preset threshold value is to set the sum of the values as the preference value,
The correlation analysis unit
A coordinate region including a coordinate axis corresponding to a weight value for a first alcoholic beverage characteristic, which is one of a plurality of preset alcoholic beverage characteristics, and a coordinate axis corresponding to a weight value for a second alcoholic beverage characteristic, which is another one of the plurality of alcoholic beverage characteristics, is generated. and a weight value for the first mainstream characteristic and a weight value for the second mainstream characteristic included in the plurality of preference information respectively matched to the plurality of different user identification codes, respectively, to each of the plurality of preference information. Placing a plurality of corresponding coordinates in the coordinate area and analyzing the correlation between the first mainstream characteristic and the second mainstream characteristic using the plurality of coordinates
Alcohol recommendation device using purchase history. delete The method of claim 1
The above alcoholic product list providing module is
a preference score calculation unit for calculating a preference score for each alcoholic product information included in the search result based on the preference information; and
Including; a liquor product list generator that generates a list of alcoholic beverages by listing a plurality of information on alcoholic beverages included in the search result in the order of preference score
Alcohol recommendation device using purchase history. According to claim 1,
an alcoholic beverage product selection unit for selecting any one alcoholic beverage product information included in the alcoholic beverage list according to a user's input;
a user location information collection unit that collects user location information from outside; and
It is located within a preset radius based on the user location information from an external DB in which liquor store information including the store location for each of a plurality of liquor stores and the amount of stock for each liquor product is stored, and corresponds to the liquor product information selected in the liquor product selection unit Alcohol recommendation device using a purchase history further comprising a; liquor store information acquisition unit for obtaining and outputting liquor store information, wherein the inventory amount of the alcoholic beverage product is greater than or equal to a preset threshold value.

Images