WO2023008617A1 - System for automatically preparing banner advertisement for internet shopping mall and method therefor - Google Patents

System for automatically preparing banner advertisement for internet shopping mall and method therefor Download PDF

Info

Publication number
WO2023008617A1
WO2023008617A1 PCT/KR2021/009930 KR2021009930W WO2023008617A1 WO 2023008617 A1 WO2023008617 A1 WO 2023008617A1 KR 2021009930 W KR2021009930 W KR 2021009930W WO 2023008617 A1 WO2023008617 A1 WO 2023008617A1
Authority
WO
WIPO (PCT)
Prior art keywords
banner
product
shopping mall
image
user
Prior art date
Application number
PCT/KR2021/009930
Other languages
French (fr)
Korean (ko)
Inventor
최윤내
Original Assignee
주식회사 아이클레이브
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 주식회사 아이클레이브 filed Critical 주식회사 아이클레이브
Publication of WO2023008617A1 publication Critical patent/WO2023008617A1/en

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241Advertisements
    • G06Q30/0251Targeted advertisements
    • G06Q30/0253During e-commerce, i.e. online transactions
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/951Indexing; Web crawling techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/20Natural language analysis
    • G06F40/258Heading extraction; Automatic titling; Numbering
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/30Semantic analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00Machine learning
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201Market modelling; Market analysis; Collecting market data
    • G06Q30/0202Market predictions or forecasting for commercial activities
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241Advertisements
    • G06Q30/0277Online advertisement
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0278Product appraisal
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/06Buying, selling or leasing transactions
    • G06Q30/0601Electronic shopping [e-shopping]
    • G06Q30/0631Item recommendations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01Social networking

Definitions

  • the present invention relates to an automatic banner advertisement creation system and method for a shopping mall on the Internet, and more particularly, to select a popular product of a shopping mall, analyze customer information of the shopping mall, classify into members and non-members, and classify as members or non-members.
  • an online shopping mall where e-commerce is conducted using a communication network such as the Internet generates banner advertisements for products that consumers want to purchase by time period, product category, and fashion, and sends the generated banner advertisement to the shopping mall. displayed on the main screen.
  • a communication network such as the Internet
  • Such banner advertisements are typically used most effectively in clothing shopping malls. For example, by season, 'Spring clothing', 'Summer clothing', 'Autumn clothing', 'Winter clothing', 'S/S (Spring/Summer)' Banners are produced for 'product exhibition', 'F/W (Fall/Winter) product exhibition', 'transitional clothing', and by category, 'outdoor product exhibition', 'men's wear discount exhibition', 'women's clothing discount exhibition', 'pants product exhibition', etc.
  • Banners are produced, and by fashion, banners are produced such as 'fashionable skirt exhibition' and ' ⁇ pattern clothing discount exhibition', and when consumers select the banner on the main screen of the shopping mall, the contents for the corresponding planned sale are displayed and consumers purchase to be able to do
  • a sales planner (Merchandiser, MD) analyzes and researches popular products by period, category, and fashion, selects popular products, selects and processes images of selected popular products, and creates banner advertisements. and displays it in the shopping mall, and if a consumer views and clicks the banner written in this way and purchases the displayed product, the purchase result is fed back, the product is reanalyzed based on the feedback result, and the product is selected according to the reanalyzed result. You decide whether or not to continue selling.
  • MD Merchandiser, MD
  • Korean Patent Publication No. 10-2016-95662 discloses an automatic advertisement banner production method and an automatic advertisement banner production server technology
  • Korean Patent Publication No. 10-2000- 7069 a method for automatically setting up a homepage associated with a free web hosting service, generating an advertisement banner, and a technology for operating the same are known.
  • the present invention has been devised to solve the above conventional problems, and provides an automatic banner advertisement creation system and method capable of automatically creating banner advertisements, which have previously been manually performed by human resources.
  • an automatic banner advertisement creation system and method for a shopping mall on the Internet of the present invention selects popular products of the shopping mall, analyzes customer information of the shopping mall, classifies them into members and non-members, It is characterized by a configuration in which a banner advertisement for a popular product selected according to members or non-members is prepared and provided to a terminal of a customer accessing a shopping mall.
  • the effect of the system and method for automatically creating banner advertisements for shopping malls on the Internet of the present invention having the above configuration is to automatically create banner advertisements for shopping malls on the Internet using artificial intelligence, so that the person in charge of the conventional shopping mall works.
  • the process of manually analyzing, selecting, and displaying products is omitted, so there is an effect that shopping mall companies can quickly create banner advertisements that match each period, category, fashion, and member status and use them for business.
  • banner advertisements which have been subjectively operated by a person in charge of a conventional shopping mall, can be customized according to popular products and member status, thereby improving customer satisfaction.
  • the quality of design and readability of the banner for products in the shopping mall is improved, and through this, it is possible to increase the shopping mall sales by improving consumers' desire to purchase.
  • FIG. 1 is a block diagram of an automatic banner advertisement creation system according to the present invention.
  • FIG. 3 is a flowchart of a method of automatically creating banner advertisements in the system of the present invention.
  • FIG. 4 is a screen of an example of a banner advertisement automatically generated according to the present invention.
  • the automatic construction system for banner advertisements for shopping malls on the Internet of the present invention is a product planner (MD, Merchandiser) in an Internet shopping mall, which is a product research, consumer analysis, product selection, display, and banner advertisement image selection. It is a system that automatically handles tasks such as processing and sales result feedback using artificial intelligence.
  • MD product planner
  • Merchandiser Merchandiser
  • a planner of a shopping mall on the Internet uses artificial intelligence to play the role of product banner and product display based on subjective judgment and automatically displays products in the shopping mall based on the sales data of the shopping mall, so that banner advertisements with high consumer attention are automatically created.
  • FIG. 1 is a block diagram of an automatic banner advertisement creation system according to the present invention.
  • the system of the present invention is a system for automatically generating a banner advertisement created based on data of a product sold in an online shopping mall that sells a specific product.
  • the above banner advertisement is created as a web page displayed on the web through the Internet or implemented as a screen displayed on a mobile phone by an application program (so-called 'app') running on a mobile communication terminal such as a user's mobile phone. do.
  • the trend collection unit 10 is a part that accesses a website of a corresponding shopping mall and extracts a keyword in order to select a popular product to be sold to the user through a banner.
  • a known crawling technique may be used to extract data by importing a web page of a shopping mall website as it is.
  • the search words extracted in this way are sorted by determining the rank from the most search word extracted the most times to the lowest search word extracted the least number of times.
  • the number of ranks is configurable.
  • the words extracted the most times are ranked from 1st to 10th in descending order of long dresses and pants. , One Piece, Cardigan, Blouse, Jacket, Skirt, Knitwear, Sweatshirt, T-shirt.
  • the trend collection unit 10 may store a keyword list including synonyms related to the corresponding product in a database for more precise search word extraction, and recognize words corresponding to the keyword as indicating the corresponding product.
  • various keywords may be set in the form of Korean, English, Korean/English mixture, English abbreviation, and Korean synonyms for the skirt product as follows.
  • skirt skirt, skirt, SK
  • Frill skirt frillskirt, frillsk, frill sk, frill skirt, frill skirt
  • Knit skirt knitskirt, knitsk, knitsk, knit skirt, knitskirt
  • Midi skirt midiskirt, midisk, midisk, midi skirt, midiskirt
  • Mini skirt miniskirt, minisk, minisk, miniskirt, miniskirt * leather skirt: leatherskirt, leathersk, leathersk, leatherskirt, leatherskirt,
  • Corduroy skirt corduroyskirt, corduroysk, corduroy sk, corduroy skirt, corduroy skirt, golden skirt, golden skirt, golden sk, golden skirt
  • the subject selector 20 selects a keyword for banner advertisement from the extracted search word group.
  • the theme selection unit 20 may select 'long dress' as a keyword for a special exhibition among them, and then a banner advertisement selling products based on the selected keyword is generated.
  • the product image collection unit 30 searches and collects the image of the product to be displayed on the banner from the image database 90 of the shopping mall based on the keyword selected by the subject selection unit 20. am.
  • the product image collection unit 30 is preferably performed using an artificial intelligence-based machine learning technique, and a classifier for extracting and classifying features from raw data
  • An image to be displayed on the banner may be searched and analyzed based on a training result by repeatedly operating the .
  • the product image collection unit 30 collects images from shopping mall websites and related SNS web pages of the same industry through a communication network when a suitable image of the corresponding product is not found in the image database 90 through a crawling technique. and can be searched and analyzed.
  • the product image collection unit 30 stores an image corresponding to the long dress in the image database 90 and a shopping mall website in the same industry. And it can be searched and analyzed by collecting from related SNS web pages.
  • the product image analysis unit 40 analyzes the characteristics of the image to be displayed on the banner collected from the product image collection unit 30, writes an appropriate phrase to be displayed on the banner, and sets a banner image with the phrase. .
  • suitable phrases to be displayed on the banner set the main title of the banner, the subtitle of the banner, and the color of the banner.
  • the main title of the banner is when the image of the product is searched and collected from the image database 90 of the shopping mall, the phrase is selected to include the product name of the product, and the image of the product is displayed on the shopping mall website and related If collected from SNS web pages, select phrases to include search terms for images.
  • the subheading of the banner is written together by selecting a phrase that matches the dictionary meaning of the word included in the main title.
  • the subtitle is accompanied by a phrase to match the long dress.
  • the color of the banner is selected to match the color of the image of the corresponding product.
  • the banner creation unit 50 defines the characteristics of the banner, such as the style, season, region, coordination, and banner schedule of the product image analyzed by the product image analysis unit 40 .
  • the data of the characteristics of the banner is stored in the banner personality database 91, and the banner creation unit 50 may create banners having various characteristics by utilizing the personality data of the banners stored in the banner personality database 91. .
  • the created banner is stored in the banner database 92.
  • styles round neck, V-neck, sweat neck, high neck, boat neck, off-shoulder, turtle neck, basic collar, high neck
  • season spring, summer, fall, winter, between seasons, etc.
  • region mountain, sea, tourist destination, amusement park, etc.
  • coordination campus, wedding, date, festival, home wear, party, company, hot place, etc.
  • banner schedule automatically creation cycle, automatic creation time, banner size, display type whether the banner is displayed on a desktop computer screen or mobile phone screen, product registration date, product analysis period, and product The nature of the banner of the price
  • Custom banner creation unit 60
  • This part extracts a list of banners suitable for the user from the banners stored in the banner database 92 and registers recommended products matching the characteristics of the extracted banners.
  • the banner may be provided to the user in various forms.
  • the product recommendation unit 70 searches the product database 94 for a product to be registered as a banner and provides the product to the customized banner creation unit 60 .
  • the product recommendation unit 70 is a product to be provided on the banner according to whether the user is a member or non-member of the shopping mall based on the user's login information among the user's experience information collected by the user experience collection unit 80 below. can determine
  • the user is a member of a shopping mall
  • popular products on sale in the shopping mall may be displayed on the banner.
  • the user is a member of a shopping mall
  • another person who has purchased related products corresponding to at least one or more of the type, brand, color, size, price range, and reviews of products purchased by the member in the past
  • products purchased by other grouped users may be recommended to members.
  • the user experience collection unit 80 receives data from the user terminal 100 such as a mobile phone, collects user experience information (login, like, shopping cart, purchase, product recommendation history information), and is linked to a database (not shown). which is stored in
  • the customized banner generating unit 60 can provide a more accurate customized product to the user through a banner by using the user experience information stored in the user experience collecting unit 80 .
  • 3 is a flow chart of a method of automatically creating a banner of the present invention.
  • the trend collection unit 10 accesses the website of the corresponding shopping mall and extracts keywords in order to select a popular product to be sold to the user through a banner.
  • a keyword to create a banner is selected from the extracted search word group.
  • the image to be displayed on the banner is searched and collected from the image database.
  • the search for the product image is preferably performed using an artificial intelligence-based machine learning technique, and learning is performed by repeatedly operating a classifier that extracts and classifies features from raw data. Based on the result of training, an image to be displayed on the banner is searched and analyzed.
  • images may be collected, searched, and analyzed from a shopping mall website of the same industry and related SNS web pages through a communication network through a crawling technique.
  • An example of the product image analysis step is the same as described in the product image analysis unit 40 described above, so detailed examples are omitted.
  • This step is to define the characteristics of the banner created in the product image analysis step (S4), such as the style, season, region, and coordination.
  • Data on the nature of the banner is stored in the banner personality database, and a plurality of banners having various characteristics can be created by utilizing the stored personality data.
  • the created banner is stored in the banner database.
  • a banner list suitable for the user is extracted from the banners stored in the banner database, and recommended products matching the characteristics of the extracted banners are registered.
  • the banner may be provided to the user in various forms by utilizing various templates stored in the banner template.
  • the recommended product is searched and provided from a product database.
  • data is received from a user terminal such as a mobile phone, and user experience information (login, like, shopping cart, purchase, product recommendation history information) is collected and stored in a linked database (not shown), and the collected user experience information is collected.
  • user experience information like, shopping cart, purchase, product recommendation history information
  • a linked database not shown
  • the collected user experience information is collected.
  • products to be provided on the banner may be determined according to whether the user is a member or non-member of the shopping mall based on the user's log-in information.
  • a main title to be displayed on the banner is set, and the main title of the banner includes the product name of the product when the image of the corresponding product is retrieved and collected from the image database of the corresponding shopping mall.
  • a phrase may be selected as much as possible, and a phrase may be selected to include a search word obtained by searching for an image when images of a corresponding product are collected from shopping mall websites and related SNS web pages in the same industry.
  • a sub-title to be displayed on the banner is set, and the sub-title of the banner may be written by selecting a phrase matching the dictionary meaning of a word included in the main title.
  • the color of the banner may be set to match the color of the product image.
  • the characteristics of the banner related to at least one or more items of the style, season, region, coordination, and banner schedule of the product image analyzed by the product image analysis unit may be defined.
  • the banner schedule defined in the banner generation step (S5) includes the automatic generation period of the banner, the automatic generation time, the size of the banner, the display type of whether the banner is displayed on a desktop computer screen or a mobile phone screen,
  • a schedule may be set for at least one item of product registration date, product analysis period, and product price.
  • the custom banner generation step (S6) when the user is a member of a shopping mall, the member corresponds to at least one or more items of the type, brand, color, size, price range, and user reviews of products purchased in the past. Related products may be displayed on the banner, and if the user is not a member of the shopping mall, popular products on sale in the shopping mall may be displayed on the banner.
  • the custom banner generation step (S6) when the user is a member of a shopping mall, the member corresponds to at least one or more items of the type, brand, color, size, price range, and user reviews of products purchased in the past. Other users who have purchased related products may be grouped, and products purchased in the corresponding group may be displayed on a banner to recommend products purchased by other grouped users to members.
  • FIG. 4 shows an example of a screen on which a banner advertisement related to 'office look' is automatically generated by the system and method of the present invention.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • Computing Systems (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Tourism & Hospitality (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Mathematical Physics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The present invention relates to a system for automatically preparing a banner advertisement for an Internet shopping mall and a method therefor and, more particularly, to a system and method for selecting a popular product of a shopping mall, classifying a member and a non-member by analyzing customer information of the shopping mall, preparing a banner advertisement of the selected popular product according to the member or the non-member, and providing the banner advertisement to a terminal of a customer who accessed the shopping mall.

Description

인터넷상의 쇼핑몰을 위한 배너 광고의 자동 작성 시스템 및 그 방법Banner advertisement automatic creation system and method for shopping malls on the Internet
본 발명은 인터넷상의 쇼핑몰을 위한 배너 광고의 자동 작성 시스템 및 그 방법에 관한 것으로서, 보다 상세하게는, 쇼핑몰의 인기상품을 선정하고, 쇼핑몰의 고객정보를 분석하여 회원 및 비회원으로 분류하고, 회원 또는 비회원에 따라서 선정된 인기상품에 대한 배너 광고를 작성하여 쇼핑몰에 접속한 고객의 단말기로 제공하는 시스템 및 그 방법에 관한 것이다.The present invention relates to an automatic banner advertisement creation system and method for a shopping mall on the Internet, and more particularly, to select a popular product of a shopping mall, analyze customer information of the shopping mall, classify into members and non-members, and classify as members or non-members. A system and method for creating a banner advertisement for a popular product selected according to non-members and providing it to a terminal of a customer accessing a shopping mall.
일반적으로, 인터넷 등의 통신망을 이용하여 전자상거래가 수행되는 온라인쇼핑몰은 해당 시기별, 상품의 카테고리별, 유행별로 소비자가 구매를 원하는 상품에 대하여 배너 광고를 생성하고, 생성된 배너 광고를 쇼핑몰의 메인 화면에 표시한다.In general, an online shopping mall where e-commerce is conducted using a communication network such as the Internet generates banner advertisements for products that consumers want to purchase by time period, product category, and fashion, and sends the generated banner advertisement to the shopping mall. displayed on the main screen.
이러한 배너 광고는 대표적으로 의류 판매용 쇼핑몰에서 가장 효과적으로 사용되고 있는데, 예를 들어, 시기별로 '봄철 의류', '여름철 의류', '가을철 의류', '겨울철 의류', 'S/S(Spring/Summer) 상품전', 'F/W(Fall/Winter) 상품전', '간절기 의류' 등으로 배너가 제작되고, 카테고리별로 '아웃도어 상품전', '남성복 할인전', '여성복 할인전', '팬츠류 상품전' 등으로 배너가 제작되고, 유행별로는 '유행 스커트 기획전', 'ㅇㅇ 무늬 의류 할인전' 등으로 배너가 제작되어, 소비자가 쇼핑몰 메인 화면에서 해당 배너를 선택하게 되면 해당 기획 판매를 위한 컨텐츠가 표시되면서 소비자가 구매를 수행할 수 있도록 한다.Such banner advertisements are typically used most effectively in clothing shopping malls. For example, by season, 'Spring clothing', 'Summer clothing', 'Autumn clothing', 'Winter clothing', 'S/S (Spring/Summer)' Banners are produced for 'product exhibition', 'F/W (Fall/Winter) product exhibition', 'transitional clothing', and by category, 'outdoor product exhibition', 'men's wear discount exhibition', 'women's clothing discount exhibition', 'pants product exhibition', etc. Banners are produced, and by fashion, banners are produced such as 'fashionable skirt exhibition' and 'ㅇㅇ pattern clothing discount exhibition', and when consumers select the banner on the main screen of the shopping mall, the contents for the corresponding planned sale are displayed and consumers purchase to be able to do
종래에는 쇼핑몰의 판매를 기획자(Merchandiser, MD)가 시기별, 카테고리별, 유행별로 인기있는 상품을 분석하고 조사하여 인기 상품을 선정하고, 선정된 인기 상품의 이미지를 선택하고 가공하여 배너광고를 작성하여 쇼핑몰에 표시하며, 이렇게 작성된 배너를 소비자가 보고 클릭하여 표시된 상품에 대한 구매가 이루어졌을 경우 구매 결과를 피드백하고, 피드백된 결과를 바탕으로 해당 상품을 재분석하고, 재분석된 결과에 따라서 해당 상품을 계속 판매할지를 결정하게 된다.Conventionally, a sales planner (Merchandiser, MD) analyzes and researches popular products by period, category, and fashion, selects popular products, selects and processes images of selected popular products, and creates banner advertisements. and displays it in the shopping mall, and if a consumer views and clicks the banner written in this way and purchases the displayed product, the purchase result is fed back, the product is reanalyzed based on the feedback result, and the product is selected according to the reanalyzed result. You decide whether or not to continue selling.
상술한 바와 같이 이루어지는 종래의 배너 광고의 작성은 관리자가 일일히 수작업으로 상품 분석 및 조사를 수행하기 때문에, 반영되는 상품들의 갯수에 한계가 있으며, 배너 작성을 위한 아이디어와 이미지 제작 및 상품의 선택과 진열에 시간이 많이 소요되는 문제점이 있었다. Since the preparation of the conventional banner advertisement made as described above is performed by the manager manually analyzing and researching the product, there is a limit to the number of products reflected, and the idea and image production for banner creation and product selection and There was a problem with the display taking a lot of time.
이로 인하여 쇼핑몰에서 배너 광고를 이용한 기획 판매를 빈번하게 실시할 수 없었으며, 기획 판매의 구매 결과를 피드백하여 반영하는 과정도 느리게 이루어질 수 밖에 없었다. Because of this, it was not possible to frequently conduct planned sales using banner advertisements in shopping malls, and the process of providing feedback and reflecting the purchase results of the planned sales was also slow.
또한, 오프라인 매장을 운영하는 판매자들은 온라인 영역으로 사업을 진출하고자 자체적으로 쇼핑몰을 제작하여 운영하고 있으나, 배너 작성을 위한 포토샵 편집기술, 웹디자인 기술 등이 부족하여 배너 작성을 외부 위탁 업체를 통하여 수행하는 실정이며 이에 따라서 쇼핑몰의 운영 비용이 증가되게 된다.In addition, sellers who operate offline stores create and operate their own shopping malls to advance into the online realm, but they lack Photoshop editing skills and web design skills to create banners, so banners are created through an external consignment company. As a result, the operating cost of the shopping mall increases.
한편, 상술한 바와 같이 인터넷상의 전자상거래를 위한 배너제작방법으로서, 대한민국 특허공개 제 10-2016-95662 호의 광고 배너 자동제작방법 및 광고배너 자동제작서버의 기술과, 대한민국특허공개 제 10-2000-7069 호의 무료 웹호스팅서비스와 연계된 홈페이지 자동셋업과 광고배너 생성방법 및 그에 관한 운용방식의 기술이 공지되어 있다.On the other hand, as described above, as a banner production method for e-commerce on the Internet, Korean Patent Publication No. 10-2016-95662 discloses an automatic advertisement banner production method and an automatic advertisement banner production server technology, and Korean Patent Publication No. 10-2000- 7069, a method for automatically setting up a homepage associated with a free web hosting service, generating an advertisement banner, and a technology for operating the same are known.
본 발명은 상기와 같은 종래의 제반 문제점을 해소하기 위하여 창안된 것으로서, 종래 인력에 의하여 수작업으로 이루어졌던 배너 광고를 자동적으로 작성할 수 있는 배너 광고의 자동 작성 시스템 및 그 방법을 제공한다.The present invention has been devised to solve the above conventional problems, and provides an automatic banner advertisement creation system and method capable of automatically creating banner advertisements, which have previously been manually performed by human resources.
상기와 같은 기술적 과제를 달성하기 위한 본 발명의 인터넷상의 쇼핑몰을 위한 배너 광고의 자동 작성 시스템 및 그 방법은, 쇼핑몰의 인기상품을 선정하고, 쇼핑몰의 고객정보를 분석하여 회원 및 비회원으로 분류하고, 회원 또는 비회원에 따라서 선정된 인기상품에 대한 배너 광고를 작성하여 쇼핑몰에 접속한 고객의 단말기로 제공하는 구성을 특징으로 한다. In order to achieve the above technical problem, an automatic banner advertisement creation system and method for a shopping mall on the Internet of the present invention selects popular products of the shopping mall, analyzes customer information of the shopping mall, classifies them into members and non-members, It is characterized by a configuration in which a banner advertisement for a popular product selected according to members or non-members is prepared and provided to a terminal of a customer accessing a shopping mall.
상기와 같은 구성을 가지는 본 발명의 인터넷상의 쇼핑몰을 위한 배너 광고의 자동 작성 시스템 및 그 방법의 효과는, 인공지능을 이용하여 인터넷상의 쇼핑몰을 위한 배너 광고를 자동적으로 작성함으로써 종래 쇼핑몰의 담당자가 일일히 수작업으로 상품을 분석하고 선정하여 진열하는 과정이 생략되어 쇼핑몰 업체가 시기별, 카테고리별, 유행별 및 회원 여부별로 부합되는 배너 광고를 신속하게 작성하여 영업에 이용할 수 있게 된 효과가 있다.The effect of the system and method for automatically creating banner advertisements for shopping malls on the Internet of the present invention having the above configuration is to automatically create banner advertisements for shopping malls on the Internet using artificial intelligence, so that the person in charge of the conventional shopping mall works. The process of manually analyzing, selecting, and displaying products is omitted, so there is an effect that shopping mall companies can quickly create banner advertisements that match each period, category, fashion, and member status and use them for business.
또한, 종래 쇼핑몰 담당자에 의하여 주관적으로 운용되었던 배너 광고가 인기상품 및 회원 여부에 따라서 맞춤형으로 제작될 수 있어 고객 만족도를 향상시킬 수 있는 효과가 있다. In addition, banner advertisements, which have been subjectively operated by a person in charge of a conventional shopping mall, can be customized according to popular products and member status, thereby improving customer satisfaction.
또한, 쇼핑몰의 상품에 대한 배너의 디자인 및 가독성의 품질이 제고되고, 이를 통하여 소비자의 구매 욕구를 향상시켜 쇼핑몰 매출의 증가를 도모할 수 있게 된다. In addition, the quality of design and readability of the banner for products in the shopping mall is improved, and through this, it is possible to increase the shopping mall sales by improving consumers' desire to purchase.
도 1 은 본 발명의 배너 광고의 자동 작성 시스템의 블럭다이어그램,1 is a block diagram of an automatic banner advertisement creation system according to the present invention;
도 2 는 본 발명의 배너 광고 자동 작성 시스템의 쇼핑트렌드 검색어 추출 을 보여주는 일례,2 is an example showing the extraction of shopping trend search words in the automatic banner advertisement creation system of the present invention;
도 3 은 본 발명의 시스템의 배너광고를 자동적으로 작성하는 방법의 순서도,3 is a flowchart of a method of automatically creating banner advertisements in the system of the present invention;
도 4 는 본 발명에 의하여 자동적으로 생성된 배너 광고의 일례의 화면이다.4 is a screen of an example of a banner advertisement automatically generated according to the present invention.
이하, 도면을 참조하여 본 발명의 인터넷상의 쇼핑몰을 위한 배너 광고의 자동 구축 시스템 및 그 방법의 구성을 설명한다.DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, configurations of a system and method for automatically constructing a banner advertisement for a shopping mall on the Internet according to the present invention will be described with reference to drawings.
단, 개시된 도면들은 당업자에게 본 발명의 사상이 충분하게 전달될 수 있도록 하기 위한 예로서 제공되는 것이다. 따라서, 본 발명은 이하 제시되는 도면들에 한정되지 않고 다른 태양으로 구체화될 수도 있다.However, the disclosed drawings are provided as examples to sufficiently convey the spirit of the present invention to those skilled in the art. Accordingly, the present invention may be embodied in other aspects without being limited to the drawings presented below.
또한, 본 발명 명세서에서 사용되는 용어에 있어서 다른 정의가 없다면, 본 발명이 속하는 기술 분야에서 통상의 지식을 가진 자가 통상적으로 이해하고 있는 의미를 가지며, 하기의 설명 및 첨부 도면에서 본 발명의 요지를 불필요하게 흐릴 수 있는 공지 기능 및 구성에 대한 상세한 설명은 생략한다.In addition, unless otherwise defined, the terms used in the specification of the present invention have meanings commonly understood by those of ordinary skill in the art to which the present invention belongs, and the gist of the present invention is described in the following description and accompanying drawings. Detailed descriptions of well-known functions and configurations that may be unnecessarily obscure are omitted.
본 발명의 인터넷상의 쇼핑몰을 위한 배너 광고의 자동 구축 시스템은, 인터넷상의 쇼핑몰에서 상품의 기획자(MD, Merchandiser)가 수행하는 업무인 상품 연구, 소비자 분석, 상품의 선택, 진열, 배너 광고의 이미지 선택과 가공, 판매 결과 피드백의 업무를 인공지능을 이용하여 자동적으로 처리하는 시스템이다.The automatic construction system for banner advertisements for shopping malls on the Internet of the present invention is a product planner (MD, Merchandiser) in an Internet shopping mall, which is a product research, consumer analysis, product selection, display, and banner advertisement image selection. It is a system that automatically handles tasks such as processing and sales result feedback using artificial intelligence.
즉, 인터넷상의 쇼핑몰의 기획자가 주관적인 판단으로 상품 배너 및 상품진열의 역할을 인공지능을 이용하여 쇼핑몰의 판매 데이터에 기초하여 자동적으로 상품을 쇼핑몰에 진열함으로써 소비자의 주목성이 높은 배너 광고가 자동으로 생성되게 된다. In other words, a planner of a shopping mall on the Internet uses artificial intelligence to play the role of product banner and product display based on subjective judgment and automatically displays products in the shopping mall based on the sales data of the shopping mall, so that banner advertisements with high consumer attention are automatically created. will be created
또한, 쇼핑몰의 회원 여부에 따라서 맞춤형으로 상품을 진열되도록 하여 매출의 향상을 도모할 수 있게 된다.In addition, it is possible to increase sales by customizing products according to whether or not they are members of the shopping mall.
도 1 은 본 발명의 배너 광고의 자동 작성 시스템의 블럭다이어그램이다.1 is a block diagram of an automatic banner advertisement creation system according to the present invention.
본 발명 시스템은, 특정 상품을 판매하는 온라인상의 쇼핑몰에서 판매되는 상품의 데이터를 기반으로 하여 작성되는 배너 광고를 자동적으로 생성하는 시스템이다.The system of the present invention is a system for automatically generating a banner advertisement created based on data of a product sold in an online shopping mall that sells a specific product.
그럼으로써, 신속하고 저렴한 비용으로 사용자의 소망에 부합되는 상품을 쇼핑몰상에 진열하고, 이를 통하여 소비자의 구매행위가 보다 더 적극적으로 이루어질 수 있게 함으로써 쇼핑몰의 매출 신장과 홍보에 기여할 수 있게 된다.In this way, it is possible to contribute to sales growth and publicity of the shopping mall by displaying products that meet the user's wishes quickly and at low cost on the shopping mall, and enabling consumers to purchase more actively through this.
상기와 같은 배너 광고는 인터넷을 통하여 웹상에서 표시되는 웹페이지로 생성되거나 또는 사용자의 휴대폰과 같은 이동통신단말기에서 실행되는 어플리케이션 프로그램(소위 '앱'(app))에 의하여 휴대폰에 표시되는 화면으로 구현된다.The above banner advertisement is created as a web page displayed on the web through the Internet or implemented as a screen displayed on a mobile phone by an application program (so-called 'app') running on a mobile communication terminal such as a user's mobile phone. do.
이하, 도면을 참조하여 본 발명의 시스템을 개조식으로 상세하게 설명하기로 한다.Hereinafter, the system of the present invention will be described in detail in a modified form with reference to the drawings.
1) 트렌드 수집부(10)1) Trend collection unit (10)
트렌드 수집부(10)는 사용자에게 배너를 통하여 판매하고자 하는 인기상품을 선정하기 위하여 해당 쇼핑몰의 웹사이트에 접속하여 검색어(Keyword)를 추출하는 부분이다.The trend collection unit 10 is a part that accesses a website of a corresponding shopping mall and extracts a keyword in order to select a popular product to be sold to the user through a banner.
이러한 상품의 검색어 추출을 위하여, 쇼핑몰 웹사이트의 웹페이지를 그대로 가져와서 데이터를 추출하는 공지의 크롤링(crawling) 기법을 활용할 수 있다.In order to extract search words for such products, a known crawling technique may be used to extract data by importing a web page of a shopping mall website as it is.
그리고, 이렇게 추출된 검색어를 가장 많은 횟수로 추출된 최다 검색어부터 가장 적은 횟수로 추출된 최저 검색어까지 순위를 정하여 정렬한다. 상기 순위의 갯수는 설정가능하다.Then, the search words extracted in this way are sorted by determining the rank from the most search word extracted the most times to the lowest search word extracted the least number of times. The number of ranks is configurable.
예를 들어, 도 2 에 도시된 바와 같이, 의류 쇼핑몰의 웹페이지를 크롤링 기법을 활용하여 검색어를 추출한 결과, 가장 많은 횟수로 추출된 단어를 내림차순으로 검색어 1위부터 10위까지가 롱원피스, 팬츠, 원피스, 가디건, 블라우스, 자켓, 스커트, 니트, 맨투맨, 티셔츠임을 나타내고 있다. For example, as shown in FIG. 2, as a result of extracting search words by using a crawling technique for a web page of a clothing shopping mall, the words extracted the most times are ranked from 1st to 10th in descending order of long dresses and pants. , One Piece, Cardigan, Blouse, Jacket, Skirt, Knitwear, Sweatshirt, T-shirt.
이때, 상기 트렌드 수집부(10)는 보다 정밀한 검색어 추출을 위하여 해당 상품에 관한 유의어들을 기재한 키워드 목록을 데이터베이스에 저장하고, 키워드에 해당하는 단어도 해당 상품을 가리키는 것으로 인식하게 할 수 있다. At this time, the trend collection unit 10 may store a keyword list including synonyms related to the corresponding product in a database for more precise search word extraction, and recognize words corresponding to the keyword as indicating the corresponding product.
예를 들어, 다음과 같이 스커트 상품에 대하여 국문, 영문, 국문 영문 혼합, 영문 약자, 한글 동의어 등의 형태로 다양한 키워드들을 설정할 수 있다.For example, various keywords may be set in the form of Korean, English, Korean/English mixture, English abbreviation, and Korean synonyms for the skirt product as follows.
* (상품명): (키워드 목록)* (product name): (keyword list)
* 스커트: skirt, 치마, SK* Skirt: skirt, skirt, SK
* 프릴스커트: frillskirt, frillsk, 프릴sk, 프릴치마, 프릴skirt* Frill skirt: frillskirt, frillsk, frill sk, frill skirt, frill skirt
* 니트스커트: knitskirt, knitsk, 니트sk, 니트치마, 니트skirt* Knit skirt: knitskirt, knitsk, knitsk, knit skirt, knitskirt
* 롱스커트: longskirt, longsk, 롱sk, 롱치마, 롱skirt* Long skirt: longskirt, longsk, longsk, long skirt, longskirt
* 미디스커트: midiskirt, midisk, 미디sk, 미디치마, 미디skirt* Midi skirt: midiskirt, midisk, midisk, midi skirt, midiskirt
* 미니스커트: miniskirt, minisk, 미니sk, 미니치마, 미니skirt * 가죽스커트: leatherskirt, leathersk, 가죽sk, 가죽치마, 가죽skirt,* Mini skirt: miniskirt, minisk, minisk, miniskirt, miniskirt * Leather skirt: leatherskirt, leathersk, leathersk, leatherskirt, leatherskirt,
레더sk, 레더스커트, 레더치마, 레더skirt * 스트라이프스커트: stripeskirt, stripesk,스트라이프sk,스트라이프치마 단가라sk, 단가라치마,단가라스커트,단가라skirtLeather sk, leather skirt, leather skirt, leather skirt
스트라이프skirt striped skirt
* 코듀로이스커트: corduroyskirt,corduroysk,코듀로이sk ,코듀로이치마, 코듀로이skirt,골덴치마,골덴스커트,골덴sk,골덴skirt* Corduroy skirt: corduroyskirt, corduroysk, corduroy sk, corduroy skirt, corduroy skirt, golden skirt, golden skirt, golden sk, golden skirt
2) 주제 선정부(20)2) Topic Selection Unit (20)
상기와 같이 트렌드수집부(10)에 의하여 쇼핑몰의 검색어가 추출되면, 주제 선정부(20)가 추출된 검색어 그룹으로부터 배너광고를 수행할 키워드를 선정한다.As described above, when the search word of the shopping mall is extracted by the trend collection unit 10, the subject selector 20 selects a keyword for banner advertisement from the extracted search word group.
예를 들어, 상술한 바와 같이, 검색어를 추출한 결과 가장 많이 등장하는 단어를 내림차순으로 검색어 1위부터 10위까지가 롱원피스, 팬츠, 원피스, 가디건, 블라우스, 자켓, 스커트, 니트, 맨투맨, 티셔츠일 경우, 주제선정부(20)가 이들중 기획전을 수행할 키워드로서 '롱원피스'를 선정할 수 있으며, 이후, 선정된 키워드에 의한 상품들을 판매하는 배너광고가 생성되게 된다.For example, as described above, as a result of extracting search words, the most frequently appearing words in descending order are long dresses, pants, dresses, cardigans, blouses, jackets, skirts, knits, sweatshirts, and T-shirts from 1st to 10th. In this case, the theme selection unit 20 may select 'long dress' as a keyword for a special exhibition among them, and then a banner advertisement selling products based on the selected keyword is generated.
3) 상품 이미지수집부(30)3) Product image collection unit (30)
상품 이미지수집부(30)는 상기 주제선정부(20)에 의하여 선정된 키워드에 기초하여, 배너(banner)에 표시될 해당 상품의 이미지를 해당 쇼핑몰의 이미지데이터베이스(90)에서 검색하여 수집하는 부분이다.The product image collection unit 30 searches and collects the image of the product to be displayed on the banner from the image database 90 of the shopping mall based on the keyword selected by the subject selection unit 20. am.
바람직하게는, 상기 상품이미지수집부(30)는 인공지능에 기초한 기계 학습기법을 이용하여 수행되는 것이 바람직하며, 원 데이터(raw data)로부터 특징적 요소(feature)들을 추출하여 분류하는 분류기(classifier)를 반복 작동시켜 학습된(training) 결과에 기초하여 배너에 표시될 이미지를 검색 및 분석할 수 있다.Preferably, the product image collection unit 30 is preferably performed using an artificial intelligence-based machine learning technique, and a classifier for extracting and classifying features from raw data An image to be displayed on the banner may be searched and analyzed based on a training result by repeatedly operating the .
상기 상품이미지수집부(30)는 이미지데이터베이스(90)에서 해당 상품의 적합한 이미지가 검색되지 아니할 경우, 통신망을 통하여 동종업계의 쇼핑몰 웹사이트 및 관련 SNS 웹페이지에서 이미지를 크롤링(crawling) 기법으로 수집하고 검색 및 분석할 수 있다. The product image collection unit 30 collects images from shopping mall websites and related SNS web pages of the same industry through a communication network when a suitable image of the corresponding product is not found in the image database 90 through a crawling technique. and can be searched and analyzed.
예를 들어, 상기 주제선정부(10)에서 키워드로서 '롱원피스'가 선정될 경우, 상품이미지수집부(30)는 롱원피스에 부합되는 이미지를 이미지데이터베이스(90)와 동종업계의 쇼핑몰 웹사이트 및 관련 SNS 웹페이지에서 수집하여 검색 및 분석할 수 있다.For example, when 'long dress' is selected as a keyword in the subject selection unit 10, the product image collection unit 30 stores an image corresponding to the long dress in the image database 90 and a shopping mall website in the same industry. And it can be searched and analyzed by collecting from related SNS web pages.
4) 상품이미지분석부(40)4) Product image analysis unit (40)
상품이미지분석부(40)는 상기 상품이미지수집부(30)로부터 수집된 배너에 표시될 이미지의 성격을 분석하여 배너에 표시될 적합한 문구를 병기하고, 문구가 병기된 배너이미지를 설정하는 부분이다.The product image analysis unit 40 analyzes the characteristics of the image to be displayed on the banner collected from the product image collection unit 30, writes an appropriate phrase to be displayed on the banner, and sets a banner image with the phrase. .
구체적으로, 상기 배너에 표시될 적합한 문구는 배너의 대제목, 배너의 소제목 및 배너의 색상을 설정한다.Specifically, suitable phrases to be displayed on the banner set the main title of the banner, the subtitle of the banner, and the color of the banner.
상기 배너의 대제목은 해당 상품의 이미지를 해당 쇼핑몰의 이미지데이터베이스(90)에서 검색하여 수집한 경우 해당 상품의 상품명이 포함되도록 문구를 선택하고, 해당 상품의 이미지를 동종업계의 쇼핑몰 웹사이트 및 관련 SNS 웹페이지에서 수집한 경우 이미지를 검색한 검색어가 포함되도록 문구를 선택한다. The main title of the banner is when the image of the product is searched and collected from the image database 90 of the shopping mall, the phrase is selected to include the product name of the product, and the image of the product is displayed on the shopping mall website and related If collected from SNS web pages, select phrases to include search terms for images.
또한, 배너의 소제목은 대제목에 포함된 단어의 사전적 의미와 부합되는 문구를 선정하여 병기하게 된다. In addition, the subheading of the banner is written together by selecting a phrase that matches the dictionary meaning of the word included in the main title.
예를 들어, 해당 상품의 상품명이 롱원피스라면, 소제목은 롱원피스와 부합되도록 문구가 병기된다. For example, if the product name of the corresponding product is a long dress, the subtitle is accompanied by a phrase to match the long dress.
문구의 일례로는 '사려면 지금이야! 스타일 좋은, 매일매일 입고 싶은, 다들 하나쯤 가지고 있는 인싸템, 인기스토어, 이 정도면, 매일 뭐 입지?, 패션의 완성은, 데일리, 편하지만 분위기 있게, 매일 아침 코디고민?, 믿고 사는, 지금 당신이 필요한, 내일 입기 좋은, 화창한 주말에 입기 좋은, 지금 계절에 딱이야, 인생컷을 책임질께, 나만의 스타일, 예쁜옷만 모았다, 예뻐서 한번 편해서 한번 더, 옷사기전 필수체크, 핫아이템, 실제로 입어보고 싶은, 코디활용하기 좋은, 스타일링 쉽고 편한' 등을 들 수 있다.An example of a phrase is 'To buy now! Good style, I want to wear it every day, Ins-item that everyone has at least one, Popular store, If this is enough, what to wear every day?, The completion of fashion, Daily, comfortable but with atmosphere, Worried about coordination every morning?, Living with faith, Now What you need, good to wear tomorrow, good to wear on a sunny weekend, perfect for the current season, I'll take care of your life cut, my own style, I've collected only pretty clothes, pretty and comfortable once more, essential checks before buying clothes, hot items, actually I want to try it on, it is easy to coordinate, and it is easy and comfortable to style.
한편, 배너의 색상은 해당 상품의 이미지가 가지고 있는 컬러와 부합되는 색상으로 선정한다.Meanwhile, the color of the banner is selected to match the color of the image of the corresponding product.
5) 배너 작성부(50)5) Banner preparation unit (50)
배너 작성부(50)는 상기 상품이미지분석부(40)에서 분석된 상품이미지의 스타일, 계절, 지역, 코디, 배너의 스케쥴 등과 같은 배너의 성격을 정의하는 부분이다.The banner creation unit 50 defines the characteristics of the banner, such as the style, season, region, coordination, and banner schedule of the product image analyzed by the product image analysis unit 40 .
상기 배너의 성격의 데이터는 배너성격데이터베이스(91)에 저장되며, 배너작성부(50)는 상기 배너성격데이터베이스(91)에 저장된 배너의 성격 데이터를 활용하여 다양한 성격을 가지는 배너를 생성할 수 있다. 생성된 배너는 배너데이터베이스(92)에 저장된다.The data of the characteristics of the banner is stored in the banner personality database 91, and the banner creation unit 50 may create banners having various characteristics by utilizing the personality data of the banners stored in the banner personality database 91. . The created banner is stored in the banner database 92.
예를 들어, 상기 상품이미지분석부(40)에서 생성된 '롱원피스'의 배너이미지에 부합되는 스타일(라운드넥,V넥, 스웨어넥, 하이넥, 보트넥, 오프숄더, 터틀넥, 베이직칼라, 하이넥칼라 등), 계절(봄, 여름, 가을, 겨울, 간절기 등), 지역(산, 바다, 관광지, 놀이공원 등), 코디(캠퍼스, 결혼식, 데이트, 페스티벌, 홈웨어, 파티, 회사, 핫플레이스, 휴가 등), 배너의 스케줄(배너의 자동 생성주기, 자동 생성시간, 배너의 사이즈, 배너가 데스크탑 컴퓨터의 화면에서 표시되는지 핸드폰의 화면에서 표시되는지의 표시유형, 상품등록일자, 상품분석기간 및 상품가격)의 배너의 성격이 정의된다. For example, styles (round neck, V-neck, sweat neck, high neck, boat neck, off-shoulder, turtle neck, basic collar, high neck) matching the banner image of the 'long dress' generated by the product image analysis unit 40 color, etc.), season (spring, summer, fall, winter, between seasons, etc.), region (mountain, sea, tourist destination, amusement park, etc.), coordination (campus, wedding, date, festival, home wear, party, company, hot place, etc.) Vacation, etc.), banner schedule (automatic banner creation cycle, automatic creation time, banner size, display type whether the banner is displayed on a desktop computer screen or mobile phone screen, product registration date, product analysis period, and product The nature of the banner of the price) is defined.
6) 맞춤형 배너 작성부(60)6) Custom banner creation unit (60)
상기 배너데이터베이스(92)에 저장된 배너로부터 사용자에게 적합한 배너의 목록을 추출하고, 추출된 배너의 성격에 부합되는 추천 상품들을 등록하는 부분이다.This part extracts a list of banners suitable for the user from the banners stored in the banner database 92 and registers recommended products matching the characteristics of the extracted banners.
이때, 배너템플레이트(template)(93)에 저장된 다양한 템플레이트를 활용하여 다양한 형태로 사용자에게 배너를 제공하여 줄 수 있다.At this time, by utilizing various templates stored in the banner template 93, the banner may be provided to the user in various forms.
또한, 상기의 추천 상품은 후술할 상품추천부(70)로부터 추천받게 된다.In addition, the above recommended products are recommended by a product recommendation unit 70 to be described later.
7) 상품 추천부(70)7) Product recommendation unit (70)
상품 추천부(70)는 배너에 등록될 상품을 상품데이터베이스(94)에서 검색하여 상기 맞춤형 배너작성부(60)에 제공하는 부분이다.The product recommendation unit 70 searches the product database 94 for a product to be registered as a banner and provides the product to the customized banner creation unit 60 .
또한, 상기 상품추천부(70)는 하기 사용자경험수집부(80)에서 수집된 사용자의 경험정보중 사용자의 로그인 정보에 기초하여 사용자가 쇼핑몰의 회원인지 비회원인지의 여부에 따라서 배너에 제공될 상품을 결정할 수 있다.In addition, the product recommendation unit 70 is a product to be provided on the banner according to whether the user is a member or non-member of the shopping mall based on the user's login information among the user's experience information collected by the user experience collection unit 80 below. can determine
일례로, 상기 사용자가 쇼핑몰의 회원일 경우에는 상기 회원이 과거에 구매하였던 상품의 종류, 브랜드, 색상, 사이즈, 가격대, 사용 후기 중 적어도 어느 하나 이상의 항목에 해당되는 관련 상품들을 배너에 표시하고, 사용자가 쇼핑몰의 비회원일 경우에는 쇼핑몰에서 판매중인 인기상품을 배너에 표시할 수 있다.For example, if the user is a member of a shopping mall, displays related products corresponding to at least one or more items of the type, brand, color, size, price range, and reviews of products purchased by the member in the past on a banner, If the user is not a member of the shopping mall, popular products on sale in the shopping mall may be displayed on the banner.
다른 일례로는, 상기 사용자가 쇼핑몰의 회원일 경우에는 상기 회원이 과거에 구매하였던 상품의 종류, 브랜드, 색상, 사이즈, 가격대, 사용 후기 중 적어도 어느 하나 이상의 항목에 해당되는 관련 상품들을 구매한 다른 사용자들을 그룹화시키고, 해당 그룹에서 구매한 상품들을 배너에 표시함으로써, 회원에게 그룹화된 다른 사용자들이 구매한 상품들을 추천하여 줄 수 있다.As another example, if the user is a member of a shopping mall, another person who has purchased related products corresponding to at least one or more of the type, brand, color, size, price range, and reviews of products purchased by the member in the past By grouping users and displaying products purchased by the corresponding group on a banner, products purchased by other grouped users may be recommended to members.
8) 사용자 경험 수집부(80)8) User experience collection unit (80)
상기 사용자경험 수집부(80)는 휴대폰과 같은 사용자단말기(100)로부터 데이터를 전달받아 사용자의 경험정보(로그인, 좋아요, 장바구니, 구매, 상품 추천이력정보)를 수집하여 연동되는 데이터베이스(미도시)에 저장하는 부분이다.The user experience collection unit 80 receives data from the user terminal 100 such as a mobile phone, collects user experience information (login, like, shopping cart, purchase, product recommendation history information), and is linked to a database (not shown). which is stored in
또한, 상기 맞춤형배너생성부(60)는 상기 사용자경험수집부(80)에 저장된 사용자 경험정보를 이용하여 사용자에게 보다 정확한 맞품형 상품을 배너를 통하여 제공하여 줄 수 있게 된다.In addition, the customized banner generating unit 60 can provide a more accurate customized product to the user through a banner by using the user experience information stored in the user experience collecting unit 80 .
다음으로, 본 발명의 배너를 자동적으로 작성하는 방법을 설명한다. 도 3 은 본 발명의 배너를 자동적으로 작성하는 방법의 순서도이다. Next, a method of automatically creating a banner according to the present invention will be described. 3 is a flow chart of a method of automatically creating a banner of the present invention.
이하, 각 단계별로 개조식으로 설명하되, 전술한 본 발명 시스템에서 설명한 부분과 동일한 구성 부분의 설명은 생략한다.Hereinafter, each step is described in a modified manner, but descriptions of the same components as those described in the system of the present invention described above are omitted.
1) 트렌드 수집단계(S1)1) Trend collection step (S1)
상기 트렌드 수집단계(S1)는 트렌드 수집부(10)는 사용자에게 배너를 통하여 판매하고자 하는 인기상품을 선정하기 위하여 해당 쇼핑몰의 웹사이트에 접속하여 검색어(Keyword)를 추출하는 부분이다.In the trend collection step (S1), the trend collection unit 10 accesses the website of the corresponding shopping mall and extracts keywords in order to select a popular product to be sold to the user through a banner.
이때, 검색어 추출시 쇼핑몰 웹사이트의 웹페이지를 그대로 가져와서 데이터를 추출하는 공지의 크롤링(crawling) 기법을 활용하는 것이 바람직하며, 이렇게 추출된 검색어를 가장 많은 횟수로 추출된 최다 검색어부터 가장 적은 횟수로 추출된 최저 검색어까지 순위를 정하여 정렬하고, 상기 순위의 갯수가 설정가능함은 전술한 트렌드 수집부(10)에서 설명한 바와 같으므로 상세 예시는 생략한다.At this time, when extracting search words, it is preferable to use a well-known crawling technique that extracts data by importing a web page of a shopping mall website as it is. Ranking is determined and sorted up to the lowest search word extracted by , and since the number of rankings can be set as described in the trend collection unit 10 described above, detailed examples are omitted.
2) 주제 선정단계(S2)2) Topic selection stage (S2)
상기 트렌드 수집단계에서 동종업체 쇼핑몰 웹사이트로부터 검색어가 추출되면, 추출된 검색어 그룹으로부터 배너를 작성할 키워드를 선정하는 단계이다. In the trend collecting step, when a search word is extracted from a shopping mall website of the same type, a keyword to create a banner is selected from the extracted search word group.
상기 주제 선정단계(S2)의 예시는 전술한 주제선정부(20)에서 설명한 바와 같으므로 상세 예시는 생략한다.Since the example of the subject selection step (S2) is the same as described in the subject selection unit 20 described above, detailed examples are omitted.
3) 상품이미지 수집단계(S3)3) Product image collection step (S3)
상기 주제 선정단계(S2)에서 선정된 키워드에 기초하여, 배너(banner)에 표시될 이미지를 이미지데이터베이스에서 검색하여 수집하는 단계이다.Based on the keyword selected in the subject selection step (S2), the image to be displayed on the banner is searched and collected from the image database.
이때, 상기 상품이미지의 검색은 인공지능에 기초한 기계 학습기법을 이용하여 수행되는 것이 바람직하며, 원 데이터(raw data)로부터 특징적 요소(feature)들을 추출하여 분류하는 분류기(classifier)를 반복 작동시켜 학습된(training) 결과에 기초하여 배너에 표시될 이미지를 검색 및 분석하게 된다.At this time, the search for the product image is preferably performed using an artificial intelligence-based machine learning technique, and learning is performed by repeatedly operating a classifier that extracts and classifies features from raw data. Based on the result of training, an image to be displayed on the banner is searched and analyzed.
이때, 이미지데이터베이스에서 적합한 이미지가 검색되지 아니할 경우에는 통신망을 통하여 동종업계의 쇼핑몰 웹사이트 및 관련 SNS 웹페이지에서 이미지를 크롤링(crawling) 기법으로 수집하고 검색 및 분석할 수 있다.At this time, when a suitable image is not found in the image database, images may be collected, searched, and analyzed from a shopping mall website of the same industry and related SNS web pages through a communication network through a crawling technique.
상기 상품이미지수집단계의 예시는 전술한 상품이미지수집부(30)에서 설명한 바와 같으므로 상세 예시는 설명한다.An example of the product image collection step is the same as described in the product image collection unit 30 described above, so a detailed example will be described.
4) 상품이미지 분석단계(S4)4) Product image analysis step (S4)
상기 상품이미지수집단계(S3)로부터 검색된 배너에 표시될 이미지에 기획전에 적합한 문구를 병기하고, 문구가 병기된 배너이미지를 생성하는 단계이다. This is a step of writing a phrase suitable for the exhibition in addition to the image to be displayed on the banner searched from the product image collection step (S3), and generating a banner image in which the phrase is written together.
상기 상품이미지분석단계의 예시는 전술한 상품이미지분석부(40)에서 설명한 바와 같으므로 상세 예시는 생략한다. An example of the product image analysis step is the same as described in the product image analysis unit 40 described above, so detailed examples are omitted.
5) 배너 생성단계(S5)5) Banner creation step (S5)
상기 상품이미지분석단계(S4)에서 생성된 배너의 스타일, 계절, 지역, 코디 등과 같은 배너의 성격을 정의하는 단계이다.This step is to define the characteristics of the banner created in the product image analysis step (S4), such as the style, season, region, and coordination.
상기 배너의 성격의 데이터는 배너성격데이터베이스에 저장되며, 저장된 성격 데이터를 활용하여 다양한 성격의 배너를 다수 생성할 수 있다. 생성된 배너는 배너데이터베이스에 저장된다.Data on the nature of the banner is stored in the banner personality database, and a plurality of banners having various characteristics can be created by utilizing the stored personality data. The created banner is stored in the banner database.
상기 배너생성단계(S5)의 예시는 전술한 배너생성부(50)에서 설명한 바와 같으므로 상세 예시는 생략한다.Since the example of the banner generating step (S5) is the same as described in the above-described banner generating unit 50, detailed examples are omitted.
6) 맞춤형 배너 생성단계(S6)6) Custom banner creation step (S6)
상기 배너생성단계(S5)에서 배너데이터베이스에 저장된 배너로부터 사용자에게 적합한 배너 목록을 추출하고, 추출된 배너의 성격에 부합되는 추천 상품들을 등록하는 단계이다.In the banner generating step (S5), a banner list suitable for the user is extracted from the banners stored in the banner database, and recommended products matching the characteristics of the extracted banners are registered.
이때, 배너템플레이트에 저장된 다양한 템플레이트를 활용하여 다양한 형태로 사용자에게 배너를 제공하여 줄 수 있다.At this time, the banner may be provided to the user in various forms by utilizing various templates stored in the banner template.
또한, 상기 추천 상품은 상품데이터베이스에서 검색하여 제공된다. In addition, the recommended product is searched and provided from a product database.
또한, 휴대폰과 같은 사용자단말기로부터 데이터를 전달받아 사용자의 경험정보(로그인, 좋아요, 장바구니, 구매, 상품 추천이력정보)를 수집하여 연동되는 데이터베이스(미도시)에 저장하고, 수집된 사용자의 경험정보중 사용자의 로그인 정보에 기초하여 사용자가 쇼핑몰의 회원인지 비회원인지의 여부에 따라서 배너에 제공될 상품을 결정할 수 있다.In addition, data is received from a user terminal such as a mobile phone, and user experience information (login, like, shopping cart, purchase, product recommendation history information) is collected and stored in a linked database (not shown), and the collected user experience information is collected. Among them, products to be provided on the banner may be determined according to whether the user is a member or non-member of the shopping mall based on the user's log-in information.
또한, 상기 상품이미지 분석단계(S4)는, 배너에 표시될 대제목을 설정하고, 상기 배너의 대제목은 해당 상품의 이미지를 해당 쇼핑몰의 이미지데이터베이스에서 검색하여 수집한 경우 해당 상품의 상품명이 포함되도록 문구를 선택하고, 해당 상품의 이미지를 동종업계의 쇼핑몰 웹사이트 및 관련 SNS 웹페이지에서 수집한 경우 이미지를 검색한 검색어가 포함되도록 문구를 선택할 수 있다.In addition, in the product image analysis step (S4), a main title to be displayed on the banner is set, and the main title of the banner includes the product name of the product when the image of the corresponding product is retrieved and collected from the image database of the corresponding shopping mall. A phrase may be selected as much as possible, and a phrase may be selected to include a search word obtained by searching for an image when images of a corresponding product are collected from shopping mall websites and related SNS web pages in the same industry.
또한, 상기 상품이미지 분석단계(S4)는, 배너에 표시될 소제목을 설정하고, 배너의 소제목은 대제목에 포함된 단어의 사전적 의미와 부합되는 문구를 선정하여 병기할 수 있다.In addition, in the product image analysis step (S4), a sub-title to be displayed on the banner is set, and the sub-title of the banner may be written by selecting a phrase matching the dictionary meaning of a word included in the main title.
또한, 상기 상품이미지 분석단계(S4)는, 해당 상품의 이미지가 가지고 있는 컬러와 어울리는 색상으로 배너의 색상을 설정할 수 있다.In addition, in the product image analysis step (S4), the color of the banner may be set to match the color of the product image.
또한, 상기 배너 생성단계(S5)는, 상기 상품이미지분석부에서 분석된 상품이미지의 스타일, 계절, 지역, 코디, 배너의 스케쥴 중 적어도 어느 하나 이상의 항목에 관한 배너의 성격을 정의할 수 있다.In addition, in the banner generation step (S5), the characteristics of the banner related to at least one or more items of the style, season, region, coordination, and banner schedule of the product image analyzed by the product image analysis unit may be defined.
또한, 상기 배너 생성단계(S5)에서 정의하는 배너의 스케줄은, 배너의 자동 생성주기, 자동 생성시간, 배너의 사이즈, 배너가 데스크탑 컴퓨터의 화면에서 표시되는지 핸드폰의 화면에서 표시되는지의 표시유형, 상품등록일자, 상품분석기간 및 상품가격 중 적어도 어느 하나 이상의 항목에 관한 스케줄을 설정할 수 있다.In addition, the banner schedule defined in the banner generation step (S5) includes the automatic generation period of the banner, the automatic generation time, the size of the banner, the display type of whether the banner is displayed on a desktop computer screen or a mobile phone screen, A schedule may be set for at least one item of product registration date, product analysis period, and product price.
또한, 상기 맞춤형 배너생성단계(S6)는, 상기 사용자가 쇼핑몰의 회원일 경우에는 상기 회원이 과거에 구매하였던 상품의 종류, 브랜드, 색상, 사이즈, 가격대, 사용 후기 중 적어도 어느 하나 이상의 항목에 해당되는 관련 상품들을 배너에 표시하고, 사용자가 쇼핑몰의 비회원일 경우에는 쇼핑몰에서 판매중인 인기상품을 배너에 표시할 수 있다.In addition, in the custom banner generation step (S6), when the user is a member of a shopping mall, the member corresponds to at least one or more items of the type, brand, color, size, price range, and user reviews of products purchased in the past. Related products may be displayed on the banner, and if the user is not a member of the shopping mall, popular products on sale in the shopping mall may be displayed on the banner.
또한, 상기 맞춤형 배너생성단계(S6)는, 상기 사용자가 쇼핑몰의 회원일 경우에는 상기 회원이 과거에 구매하였던 상품의 종류, 브랜드, 색상, 사이즈, 가격대, 사용 후기 중 적어도 어느 하나 이상의 항목에 해당되는 관련 상품들을 구매한 다른 사용자들을 그룹화시키고, 해당 그룹에서 구매한 상품들을 배너에 표시하여 회원에게 그룹화된 다른 사용자들이 구매한 상품들을 추천하여 줄 수 있다.In addition, in the custom banner generation step (S6), when the user is a member of a shopping mall, the member corresponds to at least one or more items of the type, brand, color, size, price range, and user reviews of products purchased in the past. Other users who have purchased related products may be grouped, and products purchased in the corresponding group may be displayed on a banner to recommend products purchased by other grouped users to members.
한편, 도 4 는 본 발명의 시스템 및 방법에 의하여 '오피스룩'에 관한 배너광고가 자동적으로 생성된 화면의 일례를 나타낸다.Meanwhile, FIG. 4 shows an example of a screen on which a banner advertisement related to 'office look' is automatically generated by the system and method of the present invention.
[부호의 설명][Description of code]
* 도면의 주요 부분에 대한 부호의 설명 ** Description of symbols for main parts of drawings *
10; 트렌드 수집부10; trend collection department
20; 주제 선정부 20; Subject selection department
30; 상품이미지 수집부30; Product image collection department
40; 상품이미지 분석부40; Product image analysis department
50; 배너 생성부50; banner generator
60; 맞춤형 배너 생성부60; custom banner generator
70; 상품 추천부70; Product recommendation department
80; 사용자 경험 수집부 80; User Experience Collector
90; 상품이미지 데이터베이스90; Product image database
91; 배너성격 데이터베이스91; Banner personality database
92; 배너데이터베이스92; banner database
93; 템플릿 데이터베이스 93; template database
94; 상품 데이터베이스94; product database
100; 사용자단말기100; User terminal

Claims (22)

  1. 온라인상의 쇼핑몰에서 판매되는 상품중 인기상품을 선정하고, Select popular products from among the products sold in online shopping malls,
    선정된 인기상품에 대한 배너 광고를 자동적으로 작성하여 쇼핑몰에 접속한 고객의 사용자단말기(100)로 제공하는 배너광고의 자동작성 시스템.An automatic banner advertisement creation system that automatically creates a banner advertisement for a selected popular product and provides it to a user terminal (100) of a customer accessing a shopping mall.
  2. 제 1 항에 있어서, 상기 배너광고의 자동작성 시스템은,The method of claim 1, wherein the automatic creation system of the banner advertisement,
    쇼핑몰의 인기상품을 선정하기 위하여, 해당 쇼핑몰의 웹페이지에 접속하여 판매하고자 하는 상품의 검색어를 추출하는 트렌드 수집부(10);a trend collection unit 10 that accesses a web page of the corresponding shopping mall and extracts a search term for a product to be sold in order to select a popular product of the shopping mall;
    상기 트렌드 수집부(10)에 의하여 검색어가 추출되면, 추출된 검색어 그룹으로부터 배너광고의 키워드를 선정하는 주제 선정부(20);When a search word is extracted by the trend collection unit 10, a subject selection unit 20 that selects a keyword of a banner advertisement from the extracted search word group;
    상기 주제 선정부(20)에 의하여 선정된 키워드에 기초하여 배너에 표시될 상품의 이미지를 상품이미지데이터베이스(90)에서 검색하는 상품이미지 수집부(30);a product image collection unit 30 that searches a product image database 90 for an image of a product to be displayed on a banner based on the keyword selected by the subject selection unit 20;
    상기 상품이미지 분석부(30)로부터 검색된 배너에 표시될 상품이미지에 적합한 문구를 병기하고, 문구가 병기된 배너를 생성하는 상품이미지 분석부(40);a product image analyzer 40 that writes a phrase suitable for a product image to be displayed on the banner retrieved from the product image analyzer 30 and creates a banner with the phrase written together;
    상기 상품이미지분석부(40)에서 생성된 배너의 성격을 정의하여 배너데이터베이스(92)에 저장하는 배너 작성부(50);a banner creation unit 50 defining the characteristics of the banner generated by the product image analysis unit 40 and storing them in a banner database 92;
    상기 배너데이터베이스(92)에 저장된 배너로부터 배너 목록을 추출하고, 추출된 배너의 성격에 부합되는 추천 상품들을 배너에 등록하는 맞춤형 배너생성부(60); a customized banner generator 60 that extracts a list of banners from the banners stored in the banner database 92 and registers recommended products matching the characteristics of the extracted banners with the banners;
    를 포함하여 구성되는 것을 특징으로 하는 시스템.A system characterized in that configured to include.
  3. 제 2 항에 있어서, 상기 트렌드 수집부(10)는, The method of claim 2, wherein the trend collection unit 10,
    크롤링(crawling) 기법을 이용하여 검색어를 추출하는 것을 특징으로 하는 시스템.A system characterized by extracting a search term using a crawling technique.
  4. 제 3 항에 있어서, 상기 트렌드 수집부(10)는, The method of claim 3, wherein the trend collection unit 10,
    해당 상품에 관한 유의어들을 기재한 키워드 목록을 데이터베이스에 저장하고, 키워드에 해당하는 단어도 해당 상품을 가리키는 것으로 인식하는 것을 특징으로 하는 시스템.A system characterized in that a keyword list containing synonyms related to a corresponding product is stored in a database, and words corresponding to the keywords are also recognized as indicating the corresponding product.
  5. 제 2 항에 있어서, 상기 상품이미지수집부(30)는, The method of claim 2, wherein the product image collection unit 30,
    인공지능에 기초한 기계 학습기법을 이용하여 수행되며, It is performed using machine learning techniques based on artificial intelligence,
    상품이미지데이터베이스(90)에서 적합한 이미지가 검색되지 않을 경우, 통신망을 통하여 동종업계의 쇼핑몰 웹사이트 및 관련 SNS 웹페이지에서 이미지를 수집하는 것을 특징으로 하는 시스템.When a suitable image is not found in the product image database 90, a system characterized in that for collecting images from shopping mall websites and related SNS web pages of the same industry through a communication network.
  6. 제 2 항에 있어서, According to claim 2,
    상품데이터베이스(94)에서 상품을 검색하여 상기 맞춤형 배너생성부(60)에 의하여 생성될 기획전에 제공하는 상품 추천부(70); a product recommendation unit 70 that searches for products in the product database 94 and provides them to a special exhibition to be created by the customized banner generator 60;
    사용자단말기(100)로부터 데이터를 전달받아 사용자의 경험정보를 수집하여 연동되는 데이터베이스에 저장하는 사용자경험 수집부(80); 를 더 포함하여 구성되고, a user experience collection unit 80 that receives data from the user terminal 100, collects user experience information, and stores it in a linked database; It is configured to further include,
    상기 상품추천부(70)는 하기 사용자경험수집부(80)에서 수집된 사용자의 경험정보중 사용자의 로그인 정보에 기초하여 사용자가 쇼핑몰의 회원인지 비회원인지의 여부에 따라서 배너에 제공될 상품을 결정하는 것을 특징으로 하는 시스템.The product recommendation unit 70 determines the product to be provided on the banner according to whether the user is a member or non-member of the shopping mall based on the user's login information among the user's experience information collected by the user experience collection unit 80 below. A system characterized by doing.
  7. 제 5 항에 있어서, 상기 상품이미지 분석부(40)는, The method of claim 5, wherein the product image analysis unit 40,
    배너에 표시될 대제목을 설정하고, Set the main title to be displayed on the banner,
    상기 배너의 대제목은 해당 상품의 이미지를 해당 쇼핑몰의 이미지데이터베이스(90)에서 검색하여 수집한 경우 해당 상품의 상품명이 포함되도록 문구를 선택하고, 해당 상품의 이미지를 동종업계의 쇼핑몰 웹사이트 및 관련 SNS 웹페이지에서 수집한 경우 이미지를 검색한 검색어가 포함되도록 문구를 선택하는 구성을 특징으로 하는 시스템.The main title of the banner is when the image of the product is searched and collected from the image database 90 of the shopping mall, the phrase is selected to include the product name of the product, and the image of the product is displayed on the shopping mall website and related A system characterized by a configuration for selecting a phrase to include a search term for searching for an image when collected from an SNS web page.
  8. 제 7 항에 있어서, 상기 상품이미지 분석부(40)는, The method of claim 7, wherein the product image analysis unit 40,
    배너에 표시될 소제목을 설정하고, Set the subheading to be displayed on the banner,
    배너의 소제목은 대제목에 포함된 단어의 사전적 의미와 부합되는 문구를 선정하여 병기하는 구성을 특징으로 하는 시스템.A system characterized by a configuration in which the subheading of the banner selects and writes phrases that match the dictionary meaning of words included in the main title.
  9. 제 5 항에 있어서, 상기 상품이미지 분석부(40)는, The method of claim 5, wherein the product image analysis unit 40,
    해당 상품의 이미지가 가지고 있는 컬러와 부합되는 색상으로 배너의 색상을 설정하는 것을 특징으로 하는 시스템.A system characterized in that the color of the banner is set to a color that matches the color of the image of the corresponding product.
  10. 제 1 항에 있어서, 상기 배너 작성부(50)는, The method of claim 1, wherein the banner creation unit 50,
    상기 상품이미지분석부(40)에서 분석된 상품이미지의 스타일, 계절, 지역, 코디, 배너의 스케쥴 중 적어도 어느 하나 이상의 항목에 관한 배너의 성격을 정의하는 구성을 특징으로 하는 시스템.The system characterized by a configuration for defining the nature of a banner related to at least one or more items among the style, season, region, coordination, and banner schedule of the product image analyzed by the product image analysis unit (40).
  11. 제 10 항에 있어서, 상기 배너 작성부(50)가 정의하는 배너의 스케줄은, The method of claim 10, wherein the banner schedule defined by the banner creation unit 50,
    배너의 자동 생성주기, 자동 생성시간, 배너의 사이즈, 배너가 데스크탑 컴퓨터의 화면에서 표시되는지 핸드폰의 화면에서 표시되는지의 표시유형, 상품등록일자, 상품분석기간 및 상품가격 중 적어도 어느 하나 이상의 항목에 관한 스케줄을 설정하는 것을 특징으로 하는 시스템.Automatic banner creation cycle, automatic creation time, banner size, display type of whether the banner is displayed on a desktop computer screen or mobile phone screen, product registration date, product analysis period, and product price A system characterized in that for setting a schedule for.
  12. 쇼핑몰의 배너 광고를 자동적으로 구축하기 위하여, In order to automatically build banner advertisements for shopping malls,
    사용자에게 배너를 통하여 판매하고자 하는 인기상품을 선정하기 위하여, 쇼핑몰 웹사이트에 접속하여 판매하고자 하는 상품의 검색어를 추출하는 트렌드 수집단계(S1); a trend collection step (S1) of accessing a shopping mall website and extracting search words for products to be sold in order to select popular products to be sold to users through banners;
    상기 트렌드 추출단계(S1)에서 검색어가 추출되면, 추출된 검색어 그룹으로부터 배너의 키워드를 선정하는 주제 선정단계(S2);If a search word is extracted in the trend extraction step (S1), a topic selection step (S2) of selecting a keyword of a banner from the extracted search word group;
    상기 주제 선정단계(S2)에서 선정된 키워드에 기초하여 배너에 표시될 상품이미지를 상품이미지데이터베이스에서 수집하는 상품이미지수집단계(S3);a product image collection step (S3) of collecting product images to be displayed on a banner from a product image database based on the keywords selected in the subject selection step (S2);
    상기 상품이미지수집단계(S3)에서 수집된 배너에 표시될 이미지에 적합한 문구를 병기하고, 문구가 병기된 배너를 생성하는 상품이미지 분석단계(S4);a product image analysis step (S4) of writing a phrase suitable for the image to be displayed on the banner collected in the product image collection step (S3) and generating a banner with the phrase;
    상기 상품이미지분석단계(S4)에서 생성된 배너에 기초하는 기획전의 성격을 정의하여 배너데이터베이스에 저장하는 배너 생성단계(S5);a banner creation step (S5) of defining the characteristics of a special exhibition based on the banner generated in the product image analysis step (S4) and storing them in a banner database;
    상기 배너데이터베이스에 저장된 배너로부터 배너 목록을 추출하고, 추출된 배너의 성격에 부합되는 추천 상품들을 배너에 등록하는 맞춤형 배너 생성단계(S6); a customized banner creation step (S6) of extracting a banner list from the banners stored in the banner database and registering recommended products matching the characteristics of the extracted banners in the banner;
    를 포함하여 구성되는 것을 특징으로 하는 배너 광고의 자동 작성 방법.A method of automatically creating a banner advertisement comprising:
  13. 제 12 항에 있어서, 상기 트렌드 수집단계(S1)는, The method of claim 12, wherein the trend collection step (S1),
    크롤링(crawling) 기법을 이용하여 검색어를 추출하는 것을 특징으로 하는 방법.A method characterized by extracting a search term using a crawling technique.
  14. 제 12 항에 있어서, 상기 상품이미지 수집단계(S3)는, The method of claim 12, wherein the product image collection step (S3),
    상기 이미지의 검색은 인공지능에 기초한 기계 학습기법을 이용하여 수행되는 것을 특징으로 하는 방법.The method characterized in that the search of the image is performed using a machine learning technique based on artificial intelligence.
  15. 제 12 항에 있어서, 상기 맞춤형 배너생성단계(S6)는,13. The method of claim 12, wherein the custom banner generating step (S6),
    사용자단말기로부터 데이터를 전달받아 사용자의 경험정보를 수집하여 연동되는 데이터베이스에 저장하고, 수집된 사용자의 경험정보중 사용자의 로그인 정보에 기초하여 사용자가 쇼핑몰의 회원인지 비회원인지의 여부에 따라서 배너에 제공될 상품을 결정하는 구성을 특징으로 하는 방법.Data is received from the user terminal, the user's experience information is collected, stored in the interlocking database, and based on the user's login information among the collected user's experience information, it is provided to the banner according to whether the user is a member or non-member of the shopping mall. A method of characterizing the composition that determines the product to be.
  16. 제 12 항에 있어서, 상기 상품이미지 분석단계(S4)는, The method of claim 12, wherein the product image analysis step (S4),
    배너에 표시될 대제목을 설정하고, Set the main title to be displayed on the banner,
    상기 배너의 대제목은 해당 상품의 이미지를 해당 쇼핑몰의 이미지데이터베이스에서 검색하여 수집한 경우 해당 상품의 상품명이 포함되도록 문구를 선택하고, 해당 상품의 이미지를 동종업계의 쇼핑몰 웹사이트 및 관련 SNS 웹페이지에서 수집한 경우 이미지를 검색한 검색어가 포함되도록 문구를 선택하는 구성을 특징으로 하는 방법.For the main title of the banner, when the image of the product is searched and collected from the image database of the shopping mall, the phrase is selected to include the product name of the product, and the image of the product is displayed on the shopping mall website and related SNS web pages in the same industry. A method characterized by a configuration of selecting phrases to include search terms that searched for images when collected from.
  17. 제 16 항에 있어서, 상기 상품이미지 분석단계(S4)는, The method of claim 16, wherein the product image analysis step (S4),
    배너에 표시될 소제목을 설정하고, Set the subheading to be displayed on the banner,
    배너의 소제목은 대제목에 포함된 단어의 사전적 의미와 부합되는 문구를 선정하여 병기하는 구성을 특징으로 하는 방법.The subheading of the banner is characterized by a configuration in which phrases matching the dictionary meaning of words included in the main title are selected and written together.
  18. 제 12 항에 있어서, 상기 상품이미지 분석단계(S4)는, The method of claim 12, wherein the product image analysis step (S4),
    해당 상품의 이미지가 가지고 있는 컬러와 부합되는 색상으로 배너의 색상을 설정하는 것을 특징으로 하는 방법.A method characterized in that the color of the banner is set to a color that matches the color of the image of the corresponding product.
  19. 제 12 항에 있어서, 상기 배너 생성단계(S5)는, 13. The method of claim 12, wherein the banner generating step (S5),
    상기 상품이미지분석부(40)에서 분석된 상품이미지의 스타일, 계절, 지역, 코디, 배너의 스케쥴 중 적어도 어느 하나 이상의 항목에 관한 배너의 성격을 정의하는 구성을 특징으로 하는 방법.A method characterized by a configuration for defining the characteristics of a banner related to at least one or more items of the style, season, region, coordination, and banner schedule of the product image analyzed by the product image analysis unit (40).
  20. 제 19 항에 있어서, 상기 배너 생성단계(S5)에서 정의하는 배너의 스케줄은, The method of claim 19, wherein the banner schedule defined in the banner creation step (S5),
    배너의 자동 생성주기, 자동 생성시간, 배너의 사이즈, 배너가 데스크탑 컴퓨터의 화면에서 표시되는지 핸드폰의 화면에서 표시되는지의 표시유형, 상품등록일자, 상품분석기간 및 상품가격 중 적어도 어느 하나 이상의 항목에 관한 스케줄을 설정하는 것을 특징으로 하는 방법.Automatic banner creation cycle, automatic creation time, banner size, display type of whether the banner is displayed on a desktop computer screen or mobile phone screen, product registration date, product analysis period, and product price A method characterized by setting a schedule for
  21. 제 15 항에 있어서, 상기 맞춤형 배너생성단계(S6)는,16. The method of claim 15, wherein the custom banner generating step (S6),
    상기 사용자가 쇼핑몰의 회원일 경우에는 상기 회원이 과거에 구매하였던 상품의 종류, 브랜드, 색상, 사이즈, 가격대, 사용 후기 중 적어도 어느 하나 이상의 항목에 해당되는 관련 상품들을 배너에 표시하고, If the user is a member of a shopping mall, displays related products corresponding to at least one or more of the type, brand, color, size, price range, and user reviews of products purchased by the member in the past on a banner;
    사용자가 쇼핑몰의 비회원일 경우에는 쇼핑몰에서 판매중인 인기상품을 배너에 표시하는 구성을 특징으로 하는 방법.and displaying popular products on sale in the shopping mall on a banner when the user is a non-member of the shopping mall.
  22. 제 15 항에 있어서, 상기 맞춤형 배너생성단계(S6)는,16. The method of claim 15, wherein the custom banner generating step (S6),
    상기 사용자가 쇼핑몰의 회원일 경우에는 상기 회원이 과거에 구매하였던 상품의 종류, 브랜드, 색상, 사이즈, 가격대, 사용 후기 중 적어도 어느 하나 이상의 항목에 해당되는 관련 상품들을 구매한 다른 사용자들을 그룹화시키고, 해당 그룹에서 구매한 상품들을 배너에 표시하여 회원에게 그룹화된 다른 사용자들이 구매한 상품들을 추천하여 주는 구성을 특징으로 하는 방법.If the user is a member of a shopping mall, other users who have purchased related products corresponding to at least one or more items of the type, brand, color, size, price range, and reviews of products purchased by the member in the past are grouped, A method characterized by a configuration of recommending products purchased by other grouped users to a member by displaying products purchased by the corresponding group on a banner.
PCT/KR2021/009930 2021-07-29 2021-07-29 System for automatically preparing banner advertisement for internet shopping mall and method therefor WO2023008617A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2021-0099894 2021-07-29
KR1020210099894A KR102674623B1 (en) 2021-07-29 2021-07-29 System for building the banner advertisement automatically for planning sales for shopping mall on internet and method therefor

Publications (1)

Publication Number Publication Date
WO2023008617A1 true WO2023008617A1 (en) 2023-02-02

Family

ID=85086999

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2021/009930 WO2023008617A1 (en) 2021-07-29 2021-07-29 System for automatically preparing banner advertisement for internet shopping mall and method therefor

Country Status (2)

Country Link
KR (1) KR102674623B1 (en)
WO (1) WO2023008617A1 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102663801B1 (en) * 2023-08-11 2024-05-07 머큐리프로젝트 주식회사 Method and apparatus for generating user-friendly banner advertisements in a communication system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080069955A (en) * 2008-07-21 2008-07-29 강민수 Method on advertisement using text content included in digital contents
KR101806169B1 (en) * 2016-07-25 2017-12-07 오드컨셉 주식회사 Method, apparatus, system and computer program for offering a shopping information
KR102133039B1 (en) * 2020-03-30 2020-07-10 서명교 Server for providing apparel shopping mall platform
KR20200127520A (en) * 2019-05-02 2020-11-11 주식회사 옷딜 System for building the contents automatically for planning sales and method therefor
KR20200141251A (en) * 2019-06-10 2020-12-18 오드컨셉 주식회사 Method of advertising personalized fashion item and server performing the same

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080069955A (en) * 2008-07-21 2008-07-29 강민수 Method on advertisement using text content included in digital contents
KR101806169B1 (en) * 2016-07-25 2017-12-07 오드컨셉 주식회사 Method, apparatus, system and computer program for offering a shopping information
KR20200127520A (en) * 2019-05-02 2020-11-11 주식회사 옷딜 System for building the contents automatically for planning sales and method therefor
KR20200141251A (en) * 2019-06-10 2020-12-18 오드컨셉 주식회사 Method of advertising personalized fashion item and server performing the same
KR102133039B1 (en) * 2020-03-30 2020-07-10 서명교 Server for providing apparel shopping mall platform

Also Published As

Publication number Publication date
KR102674623B1 (en) 2024-06-13
KR20230018120A (en) 2023-02-07

Similar Documents

Publication Publication Date Title
WO2020171535A2 (en) Method for providing fashion item recommendation service by using user's body shape and purchase history
WO2020085786A1 (en) Style recommendation method, device and computer program
WO2020222623A9 (en) System and method for automatically constructing content for strategic sales
WO2019156522A1 (en) Image/text-based design creating device and method
WO2020251238A1 (en) Method for obtaining user interest information on basis of input image data and method for customizing design of object
US7617016B2 (en) Computer system for rule-based clothing matching and filtering considering fit rules and fashion rules
KR102102571B1 (en) System and method for providing online shopping platform
US20200342320A1 (en) Non-binary gender filter
WO2020251174A1 (en) Method for advertising user-customized fashion item and server executing same
CN101136096A (en) Dress ornament evaluating and matching system and method on internet
WO2016153200A2 (en) Virtual fitting method and computer program
WO2018225939A1 (en) Method, device, and computer program for providing image-based advertisement
US11783408B2 (en) Computer vision based methods and systems of universal fashion ontology fashion rating and recommendation
WO2013012159A1 (en) Intelligent information providing system and method
WO2023008617A1 (en) System for automatically preparing banner advertisement for internet shopping mall and method therefor
WO2018226022A1 (en) Fashion item recommendation server, and fashion item recommendation method using same
WO2014119873A1 (en) Method for generating and identifying preferred image id code, and method for providing information using same
CN1873679A (en) System for selecting and fitting on clothes in computer network for supermarket of wearing apparel
WO2021071240A1 (en) Method, apparatus, and computer program for recommending fashion product
KR102102572B1 (en) System and method for providing online shopping mall
WO2020141802A2 (en) Method for providing fashion item recommendation service to user by using date
Aimé et al. VetiVoc: a modular ontology for the fashion, textile and clothing domain
WO2013072647A1 (en) Interactive image tagging
WO2021153964A1 (en) Fashion product recommendation method, apparatus, and system
WO2021215758A1 (en) Recommended item advertising method, apparatus, and computer program

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21951987

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE