KR102429104B1 - Product catalog automatic classification system based on artificial intelligence - Google Patents

Abstract

The present invention relates to an automatic product catalog classification system based on artificial intelligence, wherein the product catalog automatic classification system of the present invention includes a product information collection unit that receives product information, and product characteristics that analyze the product information and extract product characteristic information An extraction unit, a user information collection unit that receives user information corresponding to the product information, a user characteristic extraction unit that analyzes the collected user information to extract user characteristic information, the product characteristic information for a plurality of products, and the user By learning in advance the determination of the catalog category according to the characteristic information, information on a plurality of catalog categories is stored in a learning database, and the plurality of catalogs for the input product characteristic information and the user characteristic information by using the learning database and a product lifecycle learning unit that determines a corresponding catalog category among categories, and a catalog configuration unit that creates and stores an online store catalog and an online advertisement catalog for the catalog category determined by the product lifecycle learning unit.

Description

Product catalog automatic classification system based on artificial intelligence

The present invention relates to an automatic product catalog classification system for e-commerce, and more particularly, to an automatic product catalog classification system operated based on artificial intelligence through learning in Internet-based product sales and management.

Conventional product catalog classification is not an automatic classification method, but a method that is defined and entered by a person, and as the properties of individual products are diverse and users' tastes are diverse, most individuals such as the operator define the classification arbitrarily. .

When products are arbitrarily classified in this way, product attributes and customer expectations conflict, and products tend not to be classified in the correct classification. A problem arises in that the probability of delivering and exposing desired products to customers depends on the capabilities of individuals such as operators.

In addition, even with classification information that varies according to individual capabilities, as the types and fields of products diversify in performance management and operation prediction, there is inconsistency in product sales forecasting and operation management. have.

Accordingly, the present invention has been devised to solve the above problems, and an object of the present invention is to automatically classify product categories and catalogs based on artificial intelligence by learning user characteristics for product characteristics in Internet-based e-commerce and online It is to provide an automatic product catalog classification system that exposes catalogs for stores and catalogs for online advertising.

First, to summarize the features of the present invention, the product catalog automatic classification system according to an aspect of the present invention for achieving the above object includes: a product information collecting unit for receiving product information; a product characteristic extraction unit that analyzes the product information and extracts product characteristic information; a user information collecting unit receiving user information corresponding to the product information; a user characteristic extraction unit for extracting user characteristic information by analyzing the collected user information; By learning in advance the determination of the catalog category according to the product characteristic information and the user characteristic information for a plurality of products, the information on the plurality of catalog categories is stored in a learning database, and the product characteristic input by using the learning database a product life cycle learning unit for determining a corresponding catalog category among the plurality of catalog categories with respect to information and the user characteristic information; and a catalog configuration unit for creating an online store catalog and an online advertisement catalog for the catalog category determined by the product life cycle learning unit and storing the generated catalog in a catalog database.

The catalog for the online store includes a category for a product, product name, product description, inventory, option information, product image, included in the product characteristic information so as to correspond to the input product characteristic information and the input user characteristic information; It includes digital multimedia data for providing information on one or more of a shape, color, feature, or price as recommended information on a page of an Internet site.

The catalog for online advertisement includes a category for a product, product description, stock, option information, product image, form, included in the product characteristic information to correspond to the input product characteristic information and the input user characteristic information It includes digital multimedia data for providing information on one or more of color, feature, and price to a page of an Internet site as advertisement information.

The product life cycle learning unit, with respect to the input product characteristic information and the user characteristic information, when determining the absence of a matching catalog category in the learning database, determines the creation of a new catalog category, and adds a corresponding entry to the learning database In addition, the catalog configuration unit may generate a catalog for an online store and a catalog for an online advertisement for the new catalog category and store it in the catalog database.

The catalog configuration unit exposes the online store catalog to the logged-in user page of the Internet e-commerce system with respect to the input product characteristic information and the user characteristic information, and displays the online advertisement catalog to the logged-in user of the Internet advertisement system. can be displayed on the page.

The product life cycle learning unit determines the sales value or purchase intensity of a product with respect to the input product characteristic information and the user characteristic information, and determines the number of times the catalog category is determined by referring to the learning database, or the catalog database The exposure life period of a product is determined based on the number of times the catalog for the online store or the catalog for online advertisement obtained by referring to You can decide whether to remove the catalog or not.

The catalog configuration unit, with respect to the input product characteristic information and the user characteristic information, provides and exposes a plurality of related online store catalogs and a plurality of online advertising catalogs to the corresponding system, the sales value of the product, the purchase strength , or according to the size of the exposure life period, information on the type of exposure can be provided on the page.

The catalog configuration unit, with respect to the input product characteristic information and the user characteristic information, determines that the user characteristic information is a user who is not logged in but has the same behavior characteristics, and records the online store catalog as a non-logged-in Internet e-commerce system. It may be characterized in that it is exposed to the user's page accessed in this state, and the online advertisement catalog is exposed to the non-logged-in user's page connected to the Internet advertisement.

The catalog configuration unit, according to the exposure range determined according to the sales value of the product, the purchase strength, or the size of the exposure life period, the number of the plurality of associated online store catalogs and the associated plurality of online advertising catalogs number can be determined.

The catalog configuration unit may provide out-of-stock notice information or re-stock information from a predetermined remaining time of the exposure life period according to the information input in advance, and expose the information on the page.

And, according to another aspect of the present invention, a method for automatically classifying a product catalog in an apparatus for automatically classifying a product catalog on the Internet includes the steps of: receiving product information; extracting product characteristic information by analyzing the product information; receiving user information corresponding to the product information; extracting user characteristic information by analyzing the collected user information; Using a learning database that stores information on a plurality of catalog categories by learning in advance the determination of a catalog category according to the product characteristic information and the user characteristic information for a plurality of products, the input product characteristic information and the user determining a corresponding catalog category among the plurality of catalog categories with respect to the characteristic information; and creating a catalog for an online store and a catalog for online advertisement for the determined catalog category, and storing the generated catalog in a catalog database.

The catalog for the online store includes a category for a product, product name, product description, inventory, option information, product image, included in the product characteristic information so as to correspond to the input product characteristic information and the input user characteristic information; It includes digital multimedia data for providing information on one or more of a shape, color, feature, or price as recommended information on a page of an Internet site.

The catalog for online advertisement includes a category for a product, product description, stock, option information, product image, form, included in the product characteristic information to correspond to the input product characteristic information and the input user characteristic information It includes digital multimedia data for providing information on one or more of color, feature, and price to a page of an Internet site as advertisement information.

In the step of determining the catalog category, when determining the absence of a matching catalog category in the learning database with respect to the input product characteristic information and the user characteristic information, the creation of a new catalog category is determined, and corresponding to the learning database By adding an entry, the catalog configuration unit may create a catalog for an online store and a catalog for an online advertisement for the new catalog category and store it in the catalog database.

After the step of storing in the catalog database, with respect to the input product characteristic information and the user characteristic information, the online store catalog is exposed to the logged-in user page of the Internet e-commerce system, and the online advertisement catalog is displayed in the Internet advertising system. and exposing it to the logged-in user page of

In the step of determining the catalog category, the sales value or purchase intensity of the product is determined with respect to the input product characteristic information and the user characteristic information, and the number of times of determination of the catalog category, the catalog database with reference to the learning database The exposure life period of a product is determined based on the number of times the online store catalog or the online advertisement catalog provided by reference is obtained, and the corresponding catalog is stored in the catalog database according to the sales value, purchase intensity, or exposure life period of the product. You can decide whether to remove it or not.

After the step of storing in the catalog database, with respect to the input product characteristic information and the user characteristic information, a plurality of related online store catalogs and a plurality of online advertising catalogs are provided and exposed to the corresponding system, and the sales value of the product , providing identification information of an exposure type on the page according to the purchase intensity or the size of the exposure life period.

The method for automatically classifying the product catalog may include, according to the sales value of the product, the purchase intensity, or the size of the exposure life period, the number of associated online store catalogs and the associated plurality of online store catalogs according to a determined exposure range. The method further includes determining the number of catalogs for advertising.

After the step of storing in the catalog database, the method further includes the step of providing out-of-stock notice information or re-stocking information from the predetermined remaining time of the exposure life period according to the information input in advance and exposing the information on the page.

According to the product catalog automatic classification system according to the present invention, the performance of each product is determined and learned based on the user characteristics of the product characteristics analyzed by the metadata of the product, and product categories and catalogs are automatically classified based on artificial intelligence. It is possible to effectively expose recommended information and advertisement information according to the judgment of whether to expose the catalog for the online store and the catalog for the online advertisement, which will be exposed to , display and marketing.

The classification and composition of products according to the characteristics and life cycle of the product is ultimately related to the life cycle of the product. When it is recorded as a new product, it can be determined as a classification of a new product, and when the life cycle of a product ends, it can be determined as a classification of a discount product. As such, for the same product group, the classification judged according to the life cycle of the product is different, and this classification may affect the catalog classification and composition that can determine the purchase satisfaction of the consumer.

In the case of catalogs according to product classification in the online store, as well as information on advertisement catalogs delivered to online advertising media, the online media reduces the number of products to be exposed, thereby reducing the advertising costs consumed by advertising for products with low purchasing power. It can be reduced, and the reduced cost has utility, such as being able to invest in products that need to be advertised more intensively. In addition, it helps to circulate the company's funds through activities such as out-of-stock handling and restocking prevention for products whose life cycle is ending, and can contribute to cost reduction and operating profit increase through inventory management of products.

BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are included as part of the detailed description to help the understanding of the present invention, provide embodiments of the present invention and, together with the detailed description, explain the technical spirit of the present invention.
1 is a view for explaining an automatic product catalog classification system according to an embodiment of the present invention.
2 is a detailed block diagram of an automatic product catalog classification system according to an embodiment of the present invention.
3 is a flowchart for explaining the operation of the automatic product catalog classification system according to an embodiment of the present invention.
4 is a view for explaining an example of the implementation method of the product catalog automatic classification system according to an embodiment of the present invention.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings. In this case, the same components in each drawing are denoted by the same reference numerals as much as possible. In addition, detailed descriptions of already known functions and/or configurations will be omitted. The content disclosed below will focus on parts necessary to understand the operation according to various embodiments, and descriptions of elements that may obscure the gist of the description will be omitted. Also, some components in the drawings may be exaggerated, omitted, or schematically illustrated. The size of each component does not fully reflect the actual size, so the contents described herein are not limited by the relative size or spacing of the components drawn in each drawing.

In describing the embodiments of the present invention, if it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. And, the terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to intentions or customs of users and operators. Therefore, the definition should be made based on the content throughout this specification. The terminology used in the detailed description is for the purpose of describing embodiments of the present invention only, and should not be limiting in any way. Unless explicitly used otherwise, expressions in the singular include the meaning of the plural. In this description, expressions such as “comprising” or “comprising” are intended to indicate certain features, numbers, steps, acts, elements, some or a combination thereof, one or more other than those described. It should not be construed to exclude the presence or possibility of other features, numbers, steps, acts, elements, or any part or combination thereof.

In addition, terms such as first and second may be used to describe various components, but the components are not limited by the terms, and the terms are for the purpose of distinguishing one component from other components. used only as

I. Product catalog automatic classification system based on artificial intelligence

1 is a view for explaining an automatic product catalog classification system 100 according to an embodiment of the present invention.

Referring to FIG. 1 , an automatic product catalog classification system (or device) 100 on the Internet according to an embodiment of the present invention supports a user terminal 10 used by a buyer or customer, and online shopping for various products. devices, servers, systems, etc., such as an electronic commerce system (or server) 20, an advertisement system (or server) 30 operated by an advertiser or an advertisement agency for online advertisement, and the like, can be linked over a network.

Here, the network includes a wired and wireless Internet communication network, and further includes a network accessing the Internet through wireless communication such as WiFi and WiBro or mobile communication such as WCDMA, LTE, and 5G.

The user terminal 10 is a terminal that supports the Internet communication service, and includes a wired terminal such as a desktop PC and other communication-only terminals. can

The e-commerce system 20 corresponds to an online store that supports online shopping, and stores and manages information on products of various categories such as clothing, home appliances, food, furniture, beauty, sports, and books, along with member information. When an order is placed through the user terminal 10, delivery is made to a delivery destination through online payment. For example, there are Internet e-commerce systems operated by Auction, Gmarket, 11st Street, WeMakeup, and Timon. The e-commerce system 20 may expose and recommend information on various products based on a catalog for an online store on a page (screen) of an Internet site, and in particular, through the user terminal 10 of a logged-in user, A catalog for an online store for exposing and recommending information on a product specialized for a corresponding user on the (Internet site) page of the commerce system 20 may be managed in the database.

The advertisement system 30 is a system operated by an advertiser or an advertisement agency for online advertisement, and may be operated as an advertisement-only system on the Internet, may be a system operated within the e-commerce system 20, and may be an Internet search site ( For example, it may be a system operated by famous sites such as Daum, Naver, Google, etc.). The advertisement system 30 can advertise by exposing information about various products on the page (screen) of various Internet sites based on an online advertisement catalog for banner advertisements, etc. In particular, the user terminal 10 of the logged-in user ) through the (Internet site) page of the e-commerce system 20, Internet search sites, and sites redirected from them (eg, newspaper companies, shopping malls, etc.) Advertisement catalogs can be managed in the database.

In particular, the automatic product catalog classification system 100 on the Internet according to an embodiment of the present invention is interworking with servers on the Internet, such as the user terminal 10 , the e-commerce system 20 , and the advertisement system 30 , and the Internet In e-commerce based e-commerce, it automatically classifies product categories and catalogs based on artificial intelligence by learning user characteristics such as user feedback information on product characteristics of products managed by the e-commerce system 20, and creates catalogs for online stores and online advertisements. Effective exposure of recommended information and advertisement information was made possible on the site according to the judgment of exposure, scope, period, etc.

The product catalog automatic classification system 100 of the present invention may be operated as a dedicated system on the Internet for providing a catalog for an online store and a catalog for an online advertisement, or may be a system operated within the e-commerce system 20, It may be a system operated by a famous site such as an Internet search site (eg Daum, Naver, Google, etc.).

II. Configuration of product catalog automatic sorting system

2 is a detailed block diagram of the automatic product catalog classification system 100 according to an embodiment of the present invention.

Referring to FIG. 2 , the product catalog automatic classification system 100 according to an embodiment of the present invention includes a product information collection unit 110 , a product characteristic extraction unit 120 , a user information collection unit 130 , and user characteristics. It includes an extraction unit 140 , a product life cycle learning unit 150 , a learning database (DB) 151 , a catalog configuration unit 160 , and a catalog DB 161 .

The product information collection unit 110 receives product information. The product information is information managed for online product sales in the electronic commerce system 20 and includes product characteristic information. When the product catalog automatic classification system 100 is installed and operated in the e-commerce system 20 , the product information can be input by directly requesting and collecting product information managed in the e-commerce system 20 . Alternatively, when the product catalog automatic classification system 100 is operated separately from the e-commerce system 20, the product information may be input by requesting the e-commerce system 20 through an open API (Application Programming Interface) to collect the product information. have.

The product characteristic extraction unit 120 extracts product characteristic information by analyzing product information. The product characteristic information may include a product category, product name, product description, stock, option information, product image, shape, color, features, price, and the like.

The product information collection unit 110 requests metadata for each attribute, such as a category, product name, product description, inventory, option information, product image, shape, color, feature, and price for the product included in the product characteristic information, respectively. and transfer the received data to the product characteristic extracting unit 120 so that the product characteristic extracting unit 120 acquires the corresponding attributes of the product characteristic information one by one. It is also possible to request and receive a file format in which data is packetized, and to parse the packetized file in the product characteristic extraction unit 120 to obtain each attribute of product characteristic information.

The user information collection unit 130 receives user information corresponding to the product information received by the product information collection unit 110 . The user information is information managed to correspond with product information for online product sales in the e-commerce system 20 , and includes member information input through the user terminal 10 and feedback information input through the user terminal 10 . and user characteristic information. When the product catalog automatic classification system 100 is provided and operated in the e-commerce system 20, when requesting product information managed in the e-commerce system 20, the user information is received by requesting it together or by requesting it separately and collecting it. can Alternatively, when the product catalog automatic classification system 100 is operated separately from the e-commerce system 20, when requesting product information to the e-commerce system 20 through an open API (Application Programming Interface), request it together or separately You can also receive user information by allowing it to be collected.

The user characteristic extraction unit 140 extracts user characteristic information by analyzing the collected user information. User characteristic information includes browsing time (time looking around the product), purchase status, gender classification information, age, body type, feedback information (positive/negative opinions of product reviews or reviews), interest (interested products, whether stored in shopping carts, etc.) ), preference (product evaluation score according to purchase (asterisk, etc.)), preference (option selection information, product image/form/color/price selection information, etc.).

Since the user characteristic information can be extracted as behavior-based characteristic information, even if the user does not log in, the product corresponding to the catalog can be exposed according to the user's interest, residence time, and navigation path and type of the page.

The user information collection unit 130 requests metadata for each attribute, such as browsing time, purchase status, gender classification information, age, body type, feedback information, interest, preference, taste, etc. for the product included in the user characteristic information, respectively. The received data may be transmitted to the user characteristic extracting unit 140 so that the user characteristic extracting unit 140 acquires a corresponding attribute of the user characteristic information, and metadata of a plurality of/all attributes included in the above user characteristic information is The packetized file format may be requested and received, and the user characteristic extractor 140 may parse the packetized file to obtain each attribute of the user characteristic information.

With respect to the area in which this patent specifies users, demographic characteristics of users may not be consistent with interests. Accordingly, information such as gender and age included in the user characteristic information may not be useful in metadata. On the other hand, factors such as access area, language used, and network connection environment (type of PC/mobile device, Internet speed) may be affected.

The product life cycle learning unit 150 learns in advance the determination of a catalog category for a catalog to be created with respect to the product characteristic information from the product characteristic extraction unit 120 and the user characteristic information from the user characteristic extraction unit 140 , Information on catalog categories is stored and managed in the learning DB 151 . That is, the e-commerce system 20 learns in advance the determination of catalog categories according to product characteristic information and user characteristic information for a plurality of products of various categories managed for online sales, and learns information about the plurality of catalog categories. (151) is stored and managed.

The catalog configuration unit 160 creates an online store catalog and an online advertisement catalog for the catalog category determined by the product life cycle learning unit 150, stores and manages it in the catalog DB 161, as needed or internally/externally It can be provided upon system request.

The product life cycle learning unit 150 repeats the process of determining the catalog category with respect to the variables of each input product characteristic information and user characteristic information, according to a deep neural network algorithm, to obtain product characteristic information and user characteristic information. By accumulating the correlation information with the catalog category for the user in the learning DB 151, the catalog category for newly input product characteristic information and user characteristic information is determined consistently and reliably. Information related to the catalog category by determining the catalog category for the newly input product characteristic information and user characteristic information may also be accumulated in the learning DB 151 as one of the learning processes.

For example, the product life cycle learning unit 150, using the learning DB 151, attributes (variables) of product characteristic information A1, A2, A3, .. and attributes (variables) of user characteristic information s) With respect to B1, B2, B3,..., it is determined which catalog category corresponds to among a plurality of catalog categories accumulated in the learning DB 151 to determine the corresponding catalog category. For example, it is possible to determine an optimal catalog category for home appliances of the elderly from among a plurality of catalog categories of various types of various age groups.

A deep neural network algorithm uses an artificial neural network (ANN) including multiple hidden layers between an initial input layer and a final output layer. The multiple hidden layers of the deep neural network algorithm learn various non-linear relationships between input and output variables, and according to learning, it is possible to learn to reduce the error of correlation information with catalog categories for product characteristic information and user characteristic information. Since such a deep neural network algorithm is well known, a more detailed description will be omitted.

The catalog for the online store generated by the catalog configuration unit 160 includes a category for a product, a product name, a product description, included in the product characteristic information so as to correspond to the input product characteristic information and the input user characteristic information; It includes digital multimedia data for providing information on one or more of inventory, option information, product image, shape, color, features, and price as recommended information on a page of an Internet site.

The catalog for the online advertisement generated by the catalog configuration unit 160 is included in the product characteristic information to correspond to the input product characteristic information and the input user characteristic information, the product category, product description, stock, It includes digital multimedia data for providing information on one or more of option information, product image, shape, color, features, and price as advertisement information on a page of an Internet site.

For example, for products of the same catalog category, the catalog for an online store may include a still image form, and the catalog for an online advertisement may include a moving image or a flash image.

In addition, the product life cycle learning unit 150 determines the creation of a new catalog category when determining the absence of a matching catalog category in the learning DB 151 with respect to the newly input product characteristic information and the user characteristic information, and , the corresponding entry can be added to the learning DB 151 . Accordingly, the catalog configuration unit 160 may create an online store catalog and an online advertisement catalog for the new catalog category, store it in the catalog DB 161 and manage it. As a result of the determination of a new catalog category for the newly input product characteristic information and user characteristic information, information related to the catalog category is also accumulated in the learning DB 151 as one of the learning processes.

III. Operation of product catalog automatic sorting system

3 is a flowchart for explaining the operation of the automatic product catalog classification system 100 according to an embodiment of the present invention.

Referring to FIG. 3 , first, the product information collection unit 110 of the automatic product catalog classification system 100 receives product information ( S110 ). The product information is information managed for online product sales in the electronic commerce system 20 and includes product characteristic information.

The product characteristic extraction unit 120 analyzes product information to extract product characteristic information (S120). The product characteristic information may include a product category, product name, product description, stock, option information, product image, shape, color, features, price, and the like.

The product information collection unit 110, inside or outside the e-commerce system 20, includes categories, product names, product descriptions, stock, option information, product images, shapes, colors, and features for products included in product characteristic information. It is also possible to request metadata for each attribute, such as price, and transmit the received data to the product characteristic extraction unit 120 so that the product characteristic extraction unit 120 acquires the corresponding attributes of product characteristic information one by one. It is also possible to request and receive a file format in which metadata of a plurality/all attributes included in the product characteristic information is packetized, and parse the packetized file in the product characteristic extraction unit 120 to obtain each attribute of the product characteristic information. have.

The user information collection unit 130 receives user information corresponding to the product information received by the product information collection unit 110 ( S130 ). The user information is information managed to correspond with product information for online product sales in the e-commerce system 20 , and includes member information input through the user terminal 10 and feedback information input through the user terminal 10 . and user characteristic information.

The user characteristic extraction unit 140 extracts user characteristic information by analyzing the collected user information (S140). User characteristic information includes browsing time (time looking around the product), purchase status, gender classification information, age, body type, feedback information (positive/negative opinions of product reviews or reviews), interest (interested products, whether stored in shopping carts, etc.) ), preference (product evaluation score according to purchase (asterisk, etc.)), preference (option selection information, product image/form/color/price selection information, etc.).

The user information collection unit 130, inside or outside the e-commerce system 20, browsing time, purchase status, gender classification information, age, body type, feedback information, interest, preference, It is also possible to request metadata for each attribute, such as taste, and deliver the received data to the user characteristic extraction unit 140 so that the user characteristic extraction unit 140 acquires the corresponding attribute of the user characteristic information. It is also possible to request and receive a packetized file form in which metadata of a plurality of/all attributes included in the information is received, and the user characteristic extractor 140 parses the packetized file to obtain each attribute of the user characteristic information.

The product life cycle learning unit 150 learns in advance the determination of a catalog category for a catalog to be created with respect to the product characteristic information from the product characteristic extraction unit 120 and the user characteristic information from the user characteristic extraction unit 140 , By storing and managing information on a plurality of catalog categories in the learning DB 151, the learning DB 151 corresponds to a plurality of catalog categories for the product characteristic information and the user characteristic information input using the learning DB 151 A catalog category may be determined (S150). That is, the product life cycle learning unit 150 learns in advance the determination of a catalog category according to product characteristic information and user characteristic information for a plurality of products of various categories managed for online sales in the e-commerce system 20, Store and manage information on the catalog category of the learning DB (151) (S150).

The catalog configuration unit 160 creates an online store catalog and an online advertisement catalog for the catalog category determined by the product life cycle learning unit 150, stores and manages it in the catalog DB 161, as needed or internally/externally It may be provided according to the system's request (S160). The product life cycle learning unit 150 repeats the process of determining the catalog category with respect to the variables of each input product characteristic information and user characteristic information, according to a deep neural network algorithm, to obtain product characteristic information and user characteristic information. By accumulating the correlation information with the catalog category for the user in the learning DB 151, the catalog category for newly input product characteristic information and user characteristic information is determined consistently and reliably. Information related to the catalog category by determining the catalog category for the newly input product characteristic information and user characteristic information may also be accumulated in the learning DB 151 as one of the learning processes.

The catalog for the online store generated by the catalog configuration unit 160 includes a category for a product, a product name, a product description, included in the product characteristic information so as to correspond to the input product characteristic information and the input user characteristic information; It includes digital multimedia data for providing information on one or more of inventory, option information, product image, shape, color, features, and price as recommended information on a page of an Internet site. The catalog for the online advertisement generated by the catalog configuration unit 160 is included in the product characteristic information to correspond to the input product characteristic information and the input user characteristic information, the product category, product description, stock, It includes digital multimedia data for providing information on one or more of option information, product image, shape, color, features, and price as advertisement information on a page of an Internet site. For example, for products of the same catalog category, the catalog for an online store may include a still image form, and the catalog for an online advertisement may include a moving image or a flash image.

In addition, the catalog configuration unit 160, with respect to the input product characteristic information and the user characteristic information, includes a page on the Internet (redirected page) that the Internet e-commerce system 20 provides the catalog for the online store. ), that is, information can be provided to expose (eg, banner, flash form, etc.) to a logged-in user page, and a page on the Internet (redirected page) provided by the Internet advertising system 30 for the online advertisement catalog. included), that is, information can be provided to be exposed on the logged-in user page (eg, in the form of a banner, flash, etc. on a page of a search site or a page redirected therefrom).

In addition, the product life cycle learning unit 150 uses a predetermined function that gives weight to each variable with respect to the input product characteristic information and the user characteristic information, so that the sales value of the product (eg, a high weight to the price) ) or purchase intensity (eg, a high weight on whether to purchase). Also, the product life cycle learning unit 150 may obtain the number of times the catalog category is determined with reference to the learning DB 151 , or the catalog for the online store or the catalog for the online advertisement with reference to the catalog DB 161 . can be obtained, and based on these, the exposure life period of the product (start and end points of scheduled exposure) can be determined. Accordingly, the product life cycle learning unit 150 may determine whether to remove the corresponding catalog from the catalog DB 161 according to the sales value, purchase intensity, or exposure life period of the product. In other words, the exposure life period of individual products can be determined to be long according to the users of the main customer base, and the catalog can be appropriately removed for products with low sales value/purchase intensity, so that sales prediction and classification work can be easily performed. can do.

In addition, the catalog configuration unit 160, with respect to the input product characteristic information and the user characteristic information, a plurality of related online store catalogs and a plurality of online advertisement catalogs in the Internet e-commerce system 20, the advertisement system 30 ), etc., can provide information to be provided and exposed to the relevant system. In this case, according to the selling value of the product, the purchase intensity, or the size of the exposure life period, the exposure type classification information may be provided on the page provided by the system. For example, by providing the above classification information such as location, color, highlight, still image/video image, etc., products with high sales value and high purchase intensity are preferentially placed on the screen to increase the inflow rate and sales rate. can do.

Also, the catalog configuration unit 160 may determine the exposure range for the related catalog range according to the sales value of the product, the purchase intensity, or the size of the exposure life period. For example, the catalog configuration unit 160 determines the exposure range to be high or low, and accordingly, the number of catalogs for the plurality of online stores for the associated product characteristic/user characteristic and the plurality of the associated product characteristic/user characteristic can be determined differently for the number of catalogs for online advertising of

And, in addition, the catalog configuration unit 160, according to the information input in advance, from a predetermined remaining time (time subtracting the predetermined time from the end time) based on the exposure life period, out of stock information or restock information, etc. Accordingly, information may be provided to be displayed on a page provided by the corresponding system, such as the Internet e-commerce system 20 and the advertisement system 30 .

4 is a view for explaining an example of an implementation method of the product catalog automatic classification system 100 according to an embodiment of the present invention.

Referring to FIG. 4 , the product catalog automatic classification system 100 according to an embodiment of the present invention may be formed of hardware, software, or a combination thereof. For example, the product catalog automatic classification system 100 of the present invention may be implemented in the form of a computing system 1000 as shown in FIG. 4 or a server on the Internet having at least one processor for performing the above functions/steps/processes. can

The computing system 1000 includes at least one processor 1100 , a memory 1300 , a user interface input device 1400 , a user interface output device 1500 , a storage 1600 connected through a bus 1200 , and a network. An interface 1700 may be included. The processor 1100 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in the memory 1300 and/or the storage 1600 . The memory 1300 and the storage 1600 may include various types of volatile or nonvolatile storage media. For example, the memory 1300 may include a read only memory (ROM) 1310 and a random access memory (RAM) 1320 .

Accordingly, the steps of a method or algorithm described in connection with the embodiments disclosed herein may be directly implemented in hardware, a software module executed by the processor 1100 , or a combination of the two. A software module may be a storage/recording medium (i.e., memory 1300 and/or memory 1300) readable by a device, such as a computer, such as RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM. Alternatively, it may reside in storage 1600 . An exemplary storage medium is coupled to the processor 1100 , the processor 1100 capable of reading information from, and writing information to, the storage medium. Alternatively, the storage medium may be integrated with the processor 1100 . The processor and storage medium may reside within an application specific integrated circuit (ASIC). The ASIC may reside within the user terminal. Alternatively, the processor and storage medium may reside as separate components within the user terminal.

As described above, according to the automatic product catalog classification system 100 according to the present invention, based on user characteristics such as user feedback information on product characteristics analyzed by metadata of products managed by the e-commerce system 20, each By automatically classifying product categories and catalogs based on artificial intelligence by judging and learning the performance of products, effective recommendations for online store catalogs and online advertising catalogs that will be exposed to display and marketing according to the judgment of exposure, scope, period, etc. Information and advertisement information can be exposed. The classification and composition of products according to the characteristics and life cycle of the product is ultimately related to the life cycle of the product. When it is recorded as a new product, it can be determined as a classification of a new product, and when the life cycle of a product ends, it can be determined as a classification of a discount product. As such, for the same product group, the classification determined according to the life cycle of the product is different, and this classification may affect the catalog classification and composition that can determine the purchase satisfaction of consumers. In the case of catalogs according to product classification in the online store, as well as information on advertisement catalogs delivered to online advertising media, the online media reduces the number of products to be exposed, thereby reducing the advertising costs consumed by advertising for products with low purchasing power. It can be reduced, and the reduced cost has utility, such as being able to invest in products that need to be advertised more intensively. In addition, it helps to circulate the company's funds through activities such as out-of-stock handling and restocking prevention for products whose life cycle is ending, and can contribute to cost reduction and operating profit increase through inventory management of products.

As described above, in the present invention, specific matters such as specific components, etc., and limited embodiments and drawings have been described, but these are only provided to help a more general understanding of the present invention, and the present invention is not limited to the above embodiments. , various modifications and variations will be possible without departing from the essential characteristics of the present invention by those skilled in the art to which the present invention pertains. Accordingly, the spirit of the present invention should not be limited to the described embodiments, and all technical ideas with equivalent or equivalent modifications to the claims as well as the claims described below are included in the scope of the present invention. should be interpreted as

user terminal (10)
E-Commerce Systems (20)
Advertising system (30)
Product information collection unit 110
Product characteristic extraction unit 120
User information collection unit 130
User characteristic extraction unit 140
Product life cycle learning unit (150)
Learning DB (151)
Catalog Organizer (160)
Catalog DB(161)

Claims (20)

a product information collection unit receiving product information;
a product characteristic extraction unit that analyzes the product information and extracts product characteristic information;
a user information collecting unit receiving user information corresponding to the product information;
a user characteristic extraction unit for extracting user characteristic information by analyzing the collected user information;
By learning in advance the determination of the catalog category according to the product characteristic information and the user characteristic information for a plurality of products, the information on the plurality of catalog categories is stored in a learning database, and the product characteristic input by using the learning database a product life cycle learning unit for determining a corresponding catalog category among the plurality of catalog categories with respect to information and the user characteristic information; and
a catalog configuration unit for creating an online store catalog and an online advertisement catalog for the catalog category determined by the product life cycle learning unit and storing it in a catalog database;
including,
The product life cycle learning unit,
With respect to the input product characteristic information and the user characteristic information, when determining the absence of a matching catalog category in the learning database, it determines the creation of a new catalog category, adds a corresponding entry to the learning database, and the catalog configuration unit creating a catalog for an online store and a catalog for online advertising for the new catalog category and storing it in the catalog database;
The sales value or purchase intensity of the product is determined with respect to the inputted product characteristic information and the user characteristic information, and the number of times of determination of the catalog category obtained by referring to the learning database, or the online obtained by referring to the catalog database The exposure life period of the product exposed on the page of the Internet site is determined based on the number of times the store catalog or the online advertisement catalog is provided, and in the catalog database according to the sales value, purchase intensity, or exposure life period of the product. decides whether to remove the catalog;
The catalog configuration unit, with respect to the input product characteristic information and the user characteristic information, provides and exposes a plurality of related online store catalogs and a plurality of online advertisement catalogs to a corresponding system, the sales value of the product, the purchase strength , or provide identification information of the type of exposure on the page according to the size of the exposure life period,
The classification information is an automatic product catalog classification system, characterized in that the position, color, highlight, and whether it is a still image/moving image.
According to claim 1,
The catalog for the online store corresponds to the input product characteristic information and the user characteristic information, and includes a category for a product, product name, product description, inventory, option information, product image, and form included in the product characteristic information. Product catalog automatic classification system including digital multimedia data for providing information on one or more of , color, feature, or price as recommended information on a page of an Internet site.
According to claim 1,
The catalog for online advertisement includes product category, product description, stock, option information, product image, shape, color, A product catalog automatic classification system including digital multimedia data for providing information about one or more of features and prices as advertisement information on a page of an Internet site.

delete According to claim 1,
The catalog configuration unit exposes the online store catalog to the logged-in user page of the Internet e-commerce system with respect to the input product characteristic information and the user characteristic information, and displays the online advertisement catalog to the logged-in user of the Internet advertisement system. Product catalog automatic sorting system that exposes on the page.
According to claim 1,
The catalog configuration unit, with respect to the input product characteristic information and the user characteristic information,
The user characteristic information is identified as a user who is not logged in but whose behavior characteristics match,
A product catalog automatic classification system for exposing the online store catalog to a user page accessed while not logged in to an Internet e-commerce system, and exposing the online advertisement catalog to a non-logged user page accessing Internet advertisement.




delete delete According to claim 1,
The catalog configuration unit, according to the exposure range determined according to the sales value of the product, the purchase strength, or the size of the exposure life period, the number of the plurality of associated online store catalogs and the associated plurality of online advertising catalogs Product catalog automatic sorting system to determine the number.
According to claim 1,
The catalog configuration unit, the product catalog automatic sorting system that exposes on the page by providing out-of-stock notice information or re-stocking information from the predetermined remaining time of the exposure life period according to the information entered in advance.
In the automatic product catalog classification method in the product catalog automatic classification device on the Internet,
receiving product information;
extracting product characteristic information by analyzing the product information;
receiving user information corresponding to the product information;
extracting user characteristic information by analyzing the collected user information;
Using a learning database that stores information on a plurality of catalog categories by learning in advance the determination of a catalog category according to the product characteristic information and the user characteristic information for a plurality of products, the input product characteristic information and the user determining a corresponding catalog category among the plurality of catalog categories with respect to the characteristic information; and
Creating a catalog for an online store and a catalog for online advertising for the determined catalog category and storing it in a catalog database
With respect to the input product characteristic information and the user characteristic information, a plurality of related online store catalogs and a plurality of online advertisement catalogs are provided and exposed through the system, and the sales value, purchase intensity, or exposure life period of the product are displayed. Step of providing identification information of the type of exposure on the page according to the size
including,
In the step of determining the catalog category,
With respect to the inputted product characteristic information and the user characteristic information, when the absence of a matching catalog category is determined in the learning database, a new catalog category is created, and a corresponding entry is added to the learning database, and the catalog configuration unit includes the create a catalog for an online store and a catalog for online advertising for a new catalog category and store it in the catalog database;
The sales value or purchase intensity of a product is determined with respect to the inputted product characteristic information and the user characteristic information, the number of times of determination of the catalog category obtained by referring to the learning database, and the online store obtained by referring to the catalog database determining the exposure life period of the product based on the number of times of provision of the catalog or the online advertisement catalog, and determining whether to remove the corresponding catalog from the catalog database according to the sales value, purchase intensity, or exposure life period of the product;
The classification information is location, color, highlight, product catalog automatic classification method , characterized in that whether the still image / video image.
12. The method of claim 11,
The catalog for the online store corresponds to the input product characteristic information and the user characteristic information, and includes a category for a product, product name, product description, inventory, option information, product image, and form included in the product characteristic information. , a color, feature, or price of one or more of a product catalog automatic classification method comprising digital multimedia data for providing information on one or more of the recommended information on a page of an Internet site.
12. The method of claim 11,
The catalog for online advertisement includes product category, product description, stock, option information, product image, shape, color, A method of automatically classifying a product catalog including digital multimedia data for providing information about one or more of features and prices as advertisement information on a page of an Internet site.


delete 12. The method of claim 11,
After the step of storing in the catalog database,
exposing the online store catalog to a logged-in user page of an Internet e-commerce system with respect to the input product characteristic information and the user characteristic information, and exposing the online advertisement catalog to a logged-in user page of the Internet advertising system;
Product catalog automatic sorting method further comprising.
12. The method of claim 11,
After the step of storing in the catalog database,
With respect to the input product characteristic information and the user characteristic information, the user characteristic information is identified as a user who is not logged in but has the same behavioral characteristics,
exposing the online store catalog to a user page accessed while not logged in to an Internet e-commerce system, and exposing the online advertisement catalog to a non-logged user page accessing Internet advertisement;
Product catalog automatic sorting method further comprising.




delete delete 12. The method of claim 11,
According to the sales value of the product, the purchase intensity, or the size of the exposure life period, according to the determined exposure range, determining the number of the plurality of associated online store catalogs and the associated number of the plurality of online advertising catalogs
Product catalog automatic sorting method further comprising.
12. The method of claim 11,
After the step of storing in the catalog database,
Exposing on the page by providing out-of-stock notice information or restocking information from the predetermined remaining time of the exposure life period according to the information input in advance
Product catalog automatic sorting method further comprising.

Images