KR20160131477A - An e-commerce system based on interest category using related keywords - Google Patents

Abstract

The present invention relates to an e-commerce system based on an interest category using related keywords which classifies sale products by a provided specific purpose to form a category, and categorizes interests which are issues to classify the products by a category of an interest to provide the products by the category. The e-commerce system based on an interest category using related keywords comprises: a product registration unit to register a product and a keyword for the product; a keyword collection unit to collect a plurality of keywords, relations between the keywords, and a usage frequency of each keyword; a main category forming unit to form a highest category of an interest category using a keyword by the usage frequency of the keyword and a number of products searched by the keyword; an auxiliary category forming unit to form a lowest category for a corresponding category using related keywords related to a keyword of the corresponding category; and a product list display unit to display formed interest categories, and display products having a keyword matching a keyword of a selected category in a list if one category is selected among the interest categories. Accordingly, sale products are classified by a provided specific purpose to form a category. Specifically, interests which are issues are categorized to classify and provide the products.

Description

An e-commerce system based on interest categories using related keywords {

The present invention relates to an interest category based eCommerce system using an associated search term that categorizes articles to be classified according to a specific purpose in which sales products are provided and categorizes issues as issues and categorizes products into categories of interest .

In particular, the present invention relates to a system and method for constructing a hierarchical interest category using an associated search term extracted from a portal site or a social network, classifying a sales product according to a constructed interest category, .

In general, e-commerce sites offer classified products for sale. At this time, the category forms a hierarchical structure according to the category of the product.

For example, the top product categories are classified as fashion clothing, fashion goods, cosmetics / beauty, digital / household appliances, furniture / interior, maternity / childcare, food, sports / leisure, life / health and travel / culture. If you select the category of digital / home appliances, you will be able to select the category of digital / home appliances as cellular phones, mobile phone accessories, notebooks, PCs, monitors, cameras, games, video appliances, home appliances, sound appliances, kitchen appliances, . In such a manner, classification according to categories is subdivided into three to six steps.

The buyer goes into the detailed category step by step according to the above hierarchical category, and finally reaches the category in which the products desired by him are listed. So select the product within that category.

In addition, prior art eCommerce sites also support searches by search terms in addition to categories. Product search by search words is performed in the same way as general search engine. For example, if you enter at least one query related to a product, the products corresponding to that query are listed in its entirety.

Companies that sell products should classify products within their category by looking for categories provided by the ecommerce site, as described above.

However, vendors of products make products for specific purposes. In other words, each product is often produced and provided for a specific purpose. For example, there are products with specific purposes such as yellow sand or anti-aging. Conventional categorical classification methods do not adequately represent the specific purpose for which these products are provided.

In addition, techniques for recommending a product when purchasers want to purchase the product are analyzed by analyzing the purchase behavior of the buyer. As an example, a technique of recommending a product by simultaneously considering product introduction and evaluation information of users who purchased an actual product and preference of a user to purchase the product has been proposed. Also, based on a collaborative filtering algorithm, a technique for giving a weight to a price factor of a product is proposed as a final product recommendation [Patent Document 2].

However, the above-mentioned technologies are required to record and analyze the purchase behavior of past buyers. Also, these techniques do not reflect the specific purposes for which the seller provides the product.

[Patent Document 1] Korean Patent No. 10-1379503 (published on Mar. 28, 2014) [Patent Document 2] Korean Registered Patent No. 10-1216969 (published on Mar. 1, 2013)

The object of the present invention is to solve the above-mentioned problems, and it is an object of the present invention to construct a hierarchical interest category using an associated search term extracted from a portal site or a social network, classify sales products according to a constructed interest category , And an interest category based echarmus system using an associative search term.

In order to achieve the above object, the present invention relates to an interest category based eCommerce system using an associated search word, comprising: a product registration unit for registering a product, A main word collection unit for collecting a plurality of main words, a degree of association between the main words and a usage frequency of each main word; A main category constructing unit for constructing a top category of the interest category as a main word, and configuring a top category according to the frequency of the main word and the number of products retrieved in the main word; A sub-category constructing unit for constructing a sub-category for the category, the sub-category constructing unit comprising sub-categories with related subject words related to a subject word of the category; And a product list display unit for displaying the configured interest categories and, when one category is selected in the interest category, displaying the products having keywords matching with the keyword of the selected category in a list.

In addition, the present invention relates to an interest category based eCommerce system using an associative search term, wherein the association between the subject terms includes a direct association in which two subject terms have an association and an indirect relationship in which there is at least one subject word having a direct association in the middle And the degree of association of the indirect association is calculated as a product of the degree of association of the direct associations existing between the indirect associations.

In addition, the present invention is characterized in that the association and the frequency of use are set to a value between 0 and 1 in an interest category based eCommerce system using an associated search word.

According to another aspect of the present invention, there is provided an eCommerce system based on interest categories using an associated search term, wherein the main category constructing unit sets the top keywords as the top category according to the frequency of use of the keyword, And the number of the above-mentioned number is determined.

In addition, the present invention is characterized in that, in the e-commerce system based on the interest category using the related keyword, the product search of the corresponding keyword includes both the keyword and the related keyword associated with the keyword, .

Further, the present invention is characterized in that, in the interest category based ecolmus system using an associated keyword, the association keyword is a case where the degree of association with the keyword is greater than or equal to a predetermined minimum degree of association.

According to another aspect of the present invention, there is provided an interest category based eCommerce system using an associated keyword, wherein the degree of association is weighted by a frequency of use of an associated keyword.

According to another aspect of the present invention, there is provided an eCommerce system based on interest categories using an associative search term, the subcategory constituent unit including a keyword of one category (hereinafter referred to as a first keyword) and a related keyword of a direct association with the first keyword .

According to the present invention, in the interest category based e-commerce system using the related search word, the product list display unit also searches for products that are searched by the related keywords of the corresponding keyword and the keyword, and displays the products on a list.

According to an embodiment of the present invention, there is provided an eCommerce system based on interest categories using an associated search term, wherein the product list display unit displays a higher priority order of the keywords of the upper category of the corresponding category, In a case where the main word of the closest hierarchical level is matched, the case where the main word of the category of the closest hierarchical level is matched is displayed with a higher priority.

As described above, according to the interest category based eCommerce system using the related search word according to the present invention, the sales products can be classified according to the specific purpose in which the sales products are provided, and, in particular, It is possible to classify and provide products.

1 is a block diagram of a configuration of an overall system for implementing the present invention;
FIG. 2 is a block diagram of a configuration of an interest category based ecolmus system using an associated search term according to an embodiment of the present invention; FIG.
FIG. 3 is a graph showing main terms and associated diagrams according to an embodiment of the present invention; FIG.
FIG. 4 is a graph showing a main word and a weighted association graph according to an exemplary embodiment of the present invention; FIG.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the drawings.

In the description of the present invention, the same parts are denoted by the same reference numerals, and repetitive description thereof will be omitted.

First, an example of the overall system configuration for implementing the present invention will be described with reference to FIG.

1, the overall system for carrying out the present invention comprises a buyer terminal 11, a seller terminal 12, an e-commerce server 30, and a main language providing server 50. In addition, each terminal or server is connected through a network 20 such as the Internet, and performs data communication.

The purchaser terminal 11 is a terminal used by a purchaser and is a normal terminal having a computing function such as a PC, a notebook, a netbook, a smart phone, a tablet, and a tablet PC. The purchaser accesses the e-commerce server 30 through the purchaser terminal 11 to process the purchase operation such as retrieving the product and purchasing the retrieved product. The input data for processing is input through the input device of the buyer terminal 11 and the processing result is output through an output device such as a screen of the buyer terminal 11. [ Hereinafter, the description of the work performed by the buyer means that the work performed by the buyer terminal 11 is performed. In addition, reference numeral 11 is attached to the purchaser in addition to the terminal.

The seller terminal 12 is a terminal used by a seller, and is a normal terminal having a computing function such as a PC, a notebook, a netbook, a smart phone, a tablet, and a tablet PC. The seller accesses the e-commerce server 30 via the seller terminal 12, registers a product, and registers a keyword related to the product. Hereinafter, a description of a seller performing a job means an operation performed through the seller terminal 12. In addition, reference numeral 12 is attached to the seller in addition to the terminal.

The main keyword providing server 50 is a server that provides a main keyword, an associated keyword, an associated keyword, and the like. The keyword providing server 50 includes a search service server, a keyword analysis server, a social network server, or a social network analysis server.

The search service server is a server of a portal service company such as Naver and the following. It is a server that analyzes search terms used by users and provides related search words or real-time search word ranking. The keyword analysis server is a server that collects information on search terms of various search service servers in addition to a search service server and provides an associated search term or a real-time query ranking (or a frequency of a query real-time use).

Associated search terms are search terms that searchers use to relate to each other. For example, when a user uses search terms such as "pants" and "female pants", "men's pants", "slacks", and "skinny jeans", these search terms are analyzed do.

In addition, the real-time query ranking refers to the ranking of the frequency of use of the search terms in the current or predetermined period. In particular, the frequency (or frequency of use) of the search term for the most recent period of time is referred to as the frequency of use of the search term.

The server 50 for providing a keyword, such as a search service server, provides a frequency of using an associated keyword or a keyword.

In addition, the social network analysis server is a server that analyzes the keyword (or keyword, topic) that is an issue in the current social network and obtains frequency of use of the keyword and keyword. In addition, the social network analysis server can analyze the relevance between subject words. For example, assuming that the subject words used in a message on one social network are related to each other, the relevance or association between the subject words is obtained according to the frequency of simultaneous display in the message.

The main language providing server 50 provides a search term or a main term, a degree of association between them, and a usage frequency (real-time rank, etc.) of each term or keyword, such as a search service server or a social network analysis server. Hereinafter, for convenience of explanation, the search terms or main terms are all referred to as main words, and the relevance between the main terms is referred to as a relation, and the real time ranking or frequency of use of the main words is used as the frequency of use of the main word. That is, the main language providing server 50 provides the main language, the degree of association between the main words, and the usage frequency of each main language.

Next, the e-commerce server 30 is a normal web server, which is a server that provides sales and purchase services of products on the web. The e-commerce server 30 may be configured as a web server providing a main interface related to the e-commerce service on the web, or a WAS server providing various applications required in the e-commerce interface of the web server.

The ecommerce server 30 receives the main word from the main word providing server 50, the degree of association between the main words, and the frequency of use of each main word, and provides the category of the product by the main word. At this time, the category consisting of subjects is called the interest category.

The database 40 includes a product DB 41 for storing information on the product, a main word DB 42 for storing information related to the main word, and a category DB 43 for storing interest categories classified into main words do. In addition, a user DB for storing user information and the like may be added. The configuration of the database 40 is only a preferred embodiment. In the development of a specific device, the database 40 may have a different structure according to the database construction theory in consideration of the ease and efficiency of access and search.

Next, the detailed configuration of the interest category based eCommerce system 30 using an associated search term according to an embodiment of the present invention will be described in more detail with reference to FIG.

2, the eCommerce system 30 according to the present invention includes a product registration unit 31 for registering a sales product, a main word collection unit 32 for collecting information on the main word, A main category construction unit 33, a sub category construction unit 34 constituting a detailed category, and a product list display unit 35 displaying products of the category. And a product transaction processing unit 36 for trading sales products.

The product registration unit 31 receives the information of the product to be sold from the seller terminal 12, and registers the product to be sold. At this time, the product registration unit 31 receives the keywords for the sales product or constructs keywords of the sales product using the inputted keywords.

For example, when a seller registers a product for a yellow sand mask, a keyword such as "yellow sand "," mask " At this time, the product registration unit 31 can add "fine dust", which is a similar keyword to "yellow sand", and register it.

The product registration unit 31 can extract keywords from the product information. For example, the product size may be set as a keyword, for example, "child" if it is for a child, "woman" or "old person" for a woman or an elderly person.

Next, the main word collection unit 32 connects to the main word providing server 50 and collects information about the main word. Preferably, the main word, the degree of association between the main words, and the frequency of use of the main word are collected.

Next, the main word collection unit 32 normalizes the association between the main words or the usage frequency of the main word.

Preferably, the degree of association is scored between 0 and 1 to normalize. At this time, the larger the association, the greater the relevance between subject words. The association degree 0 indicates that there is no relation to each other at all.

Fig. 3 illustrates the relationship between the main word and the main word. As shown in Fig. 3, the main words are represented by T1, T2, ..., T9, and the degree of association between the main words is indicated by a score.

On the other hand, in FIG. 3, main terms T1 and T8 have no direct relationship. However, the main word T1 is directly related to the main word T2, and the main word T2 is directly related to the main word T8. In this case, the degree of association is calculated by multiplying the indirect association between the main words T1 and T8. That is, the correlation degree 0.9 of T1-T2 and the correlation degree 0.7 of T2-T8 are multiplied. The result of multiplying these associations is 0.63 (= 0.9 × 0.7), which is an association of the indirect association T1-T7.

In addition, the degree of association between the two subjects is determined by associating the highest score among the direct association or indirect association. In the example of FIG. 3, the association between the main words T1 and T3 has a direct association of 0.4 and the indirect association (T1-T5-T7) of 0.56 (= 0.8 x 0.7). Therefore, the association between T1 and T3 is 0.56.

On the other hand, the frequency of use of the main word is also normalized to a score between 0 and 1. The larger the frequency of use, the more frequently the subject is used. In addition, a high frequency of use of a keyword indicates that the keyword is hot.

Next, the main category constructing unit 33 constitutes the interest category as the main word. At this time, a category is formed by the frequency of use of the main word and the number of products to be searched in the main word.

Preferably, the higher-level keywords are set as categories according to the frequency of use of the keyword, but only when the number of products retrieved by the corresponding keyword is greater than or equal to a predetermined number. More preferably, the search by the main word includes a search by a related keyword (hereinafter referred to as an associated keyword) in addition to the main keyword. In particular, the search range of the related keyword is determined by the degree of association with the keyword.

For example, if it is ranked in the order of "National Pension Reform", "Sohn Heungmin", "Yellow Sand", "Moon Jae In", "Kim Yeon A" Search for products. If the search results for each subject are 10, 120, 500, 30, 200, etc., and the minimum number of searches is 50, only "Sohn Heungmin", "Yellow Sand", " Is selected. The subject term consists of interest categories (or issue categories).

The number of interest categories can also be predetermined. For example, if the number of interest categories is set to 8, an upper number of subject words are selected up to 8, and the selected subject words are set as interest categories.

Next, a first method of setting a search range for a main word will be described. As described above, the search range is a related keyword associated with a corresponding keyword and a corresponding keyword. At this time, the minimum criterion correlation is set to 0.5. The minimum criterion correlation or minimum correlation represents the minimum correlation of the relevant keyword to be included in the search scope.

3, it is assumed that the subject word is T1. T5, T6, T8, and T7 are associated with T1 with a relevance of 0.5 or greater. Each has a correlation of 0.9, 0.56, 0.8, 0.56, 0.63, 0.567. In particular, the main word T3 is an indirect correlation by T1-T5-T3, which is defined as a product of 0.8 × 0.7. The main word T7 is the correlation of the indirect association by T1-T2-T8-T7, which is defined as the product of 0.9 × 0.7 × 0.9. If there are more than 50 product search results with the keyword T1 and the related keywords T2, T3, T5, T6, T8, and T7 as keywords, the keyword T1 is selected as the category.

Next, a second method of setting the search range for the main word will be described. This method is the same as the first method described above, but calculates a weighted association to which a weight is assigned. At this time, the weight is determined by the frequency of use of the related keyword. That is, if the frequency of use of the related keyword is high, the weight is increased, and if the frequency of use of the related keyword is low, the weight is given low.

FIG. 4 is a graph showing the degree of association given weight according to frequency of use in the association diagram of FIG. 3. FIG. In FIG. 4, it can be seen that T1-T3 has a small degree of association but a high frequency of use, so that its weight is increased to 0.6, T1-T5 has a relatively high association degree, but its frequency is low and its weight is reduced to 0.6. As a result, the degree of association between T1-T3 and T1-T5 was equally 0.6.

At this time, the degree of association is weighted by the frequency of use of the corresponding keyword in the main word T1. Weights are given by the frequency of use of T1, the relative subject of T1, in T3. The target subject is the farther subject in the subject. That is, the frequency of use of T1 is not determined. In the relationship of T1-T5-T3, the weight for the association of T1-T5 is determined by the frequency of use of T5, and the weight for the association of T5-T3 is determined by the frequency of use of T3. Thus, in the indirect association of T1-T5-T3 and the indirect association of T1-T3-T5, T3-T5 have the same association but different weighting associations with weights. That is, the former is weighted by the frequency of use of T3, and the latter is weighted by the frequency of use of T5.

Weighting associations with frequency of use is aimed at raising the prioritized subject words.

Next, the sub-category constructing unit 34 forms a subcategory for the main word of the category. At this time, the related keywords having a direct relation with the keyword of the category are selected as the keyword of the subcategory.

In the example of FIG. 3, it is assumed that the main word of the current category is T1. At this time, the main subjects which are directly related to the main word T1 are T2, T3, T5, T8. Therefore, each of these four main words is composed of subcategories. Therefore, when the buyer selects the category T1, four categories T2, T3, T5 and T8 are set as subcategories of the main word T1.

At this time, the minimum value of the degree of association may be determined and limited to a category having a degree of association of a minimum value or more. It can also be limited by the number of product searches by subject keywords in the subcategories. The method of limiting the search range by the number of product searches is the same as the previous method of selecting the main category.

On the other hand, the sub-category constructing unit 34 may also assign a weight according to the frequency of use to the degree of association.

Next, the product list display unit 35 displays the product lists searched by the main word of the category. At this time, if the keyword of the sales product matches the corresponding keyword, the product is searched as a product in the category.

Preferably, the product list display unit 35 also searches for products that are searched by the corresponding keyword and an associated keyword of the keyword, and displays the list on the list. The related keyword can be extracted by the association graph and weighted using the frequency of use can be given to the association graph.

Also, the product list display unit 35 increases the priority of the keywords of the upper category of the corresponding category as the keywords of the upper category of the corresponding category are displayed.

For example, in the example of FIG. 3, when the interest category is set in the order of T1-T2-T7 and descended into the subcategory, the product in which all the three main words are matched is placed at the top in the product list, Are placed on top of each other.

In addition, when the same number of subject words are matched, priority is given to the case where the main word of the category of the nearest layer matches. That is, if there is a sales product A that matches both T1 and T2 and a sales product B that matches both T1 and T3, the latter, product B, becomes higher priority.

Next, the product transaction processing unit 36 performs a work on product sales when a product is selected from a product list and a purchase request is made. The product transaction processing unit 36 includes a module for creating a product purchase page and electronic payment, and performs a normal online product purchase or sale operation.

The invention made by the present inventors has been described concretely with reference to the embodiments. However, it is needless to say that the present invention is not limited to the embodiments, and that various changes can be made without departing from the gist of the present invention.

11: buyer terminal 12: seller terminal
20: network 30: ecommerce server
31: product registration part 32: main word collection part
33: main category construction part 34: sub category construction part
35: product list display section 36: product transaction processing section
40: Database 41: Product DB
42: Keyword DB 43: Category DB
50:

Claims (10)

In an interest category based e-commerce system using associative search terms,
A product registration unit for registering a product, the product registration unit registering a keyword for the product;
A main word collection unit for collecting a plurality of main words, a degree of association between the main words and a usage frequency of each main word;
A main category constructing unit for constructing a top category of the interest category as a main word, and configuring a top category according to the frequency of the main word and the number of products retrieved in the main word;
A sub-category constructing unit for constructing a sub-category for the category, the sub-category constructing unit comprising sub-categories with related subject words related to a subject word of the category; And
And a product list display unit for displaying a list of products having a keyword that matches the keyword of the selected category when one category is selected in the interest category. Based ecommerce system
The method according to claim 1,
Wherein the association between the subject terms includes a direct association in which two subject terms have an association and an indirect association in which there is at least one subject word having a direct association in between,
Wherein the degree of association of the indirect association is calculated as a product of the degree of association of the direct associations existing between the indirect associations.
3. The method of claim 2,
Wherein the association and the use frequency are set to a value between 0 and 1. The interest category based eCommerce system using an associated search term.
3. The method of claim 2,
Wherein the main category constructing unit defines the top keywords as the top category according to the frequency of use of the keyword and defines only when the number of products searched for by the keyword is equal to or greater than a predetermined number. Mus systems.
5. The method of claim 4,
Wherein the product search of the subject keyword includes both a subject keyword and a product keyword based on an associated keyword that is related to the subject keyword, based on the interest category based eCommerce system.
The method according to claim 6,
Wherein the related keyword is a case where the degree of association with the corresponding keyword is equal to or greater than a predetermined minimum degree of association.
The method according to claim 1,
Wherein the relevance is weighted by a frequency of use of an associated main word.
3. The method of claim 2,
Wherein the sub-category constructing unit selects a category keyword (hereinafter referred to as a first keyword) and a related keyword having a direct association with the first keyword as a sub-category.
3. The method of claim 2,
Wherein the product list display unit also searches for products that are searched for by the corresponding keyword and a related keyword of the keyword, and displays the selected keyword on a list.
10. The method of claim 9,
The product list display unit displays the higher-priority keywords of the upper category of the corresponding category with higher priority as they are matched with the keywords of the corresponding product, and when the same number of the keyword words are matched, Based on the interest category based on the related keyword.

Images