KR20200121391A - Open market big data based AI keyword recommending system and method - Google Patents

Abstract

Disclosed are an open market big data-based artificial intelligence (AI) keyword recommendation system and method. According to one embodiment of the present invention, the system for recommending a keyword according to AI utilizing open market big data may comprise: a big data collection engine collecting data on a customer movement path for a customer shopping in open market platform; a big data analysis module analyzing the data collected by the big data collection engine to perform keyword analysis by seller product; and a customer channel module recommending an exposure optimization keyword by seller product based on the keyword analysis result.

Description

Open market big data based AI keyword recommending system and method}

The present invention relates to an open market big data based AI keyword recommendation system and method.

Recently, in the e-commerce market, artificial intelligence is provided to consumers through the AI open market in consumer-oriented simple search product purchase such as image search, product tagging, and application of big data product recommendations based on personal search history at the level of existing strong search base and price comparison. It is changing to a pattern that automatically recommends customized products.

AI open market technology is analyzed based on big data of various search histories such as user's pre-consumption exploration and purchase, and the user's search keyword and product consumption pattern data are the key data. In this personalized AI open market market, it has become important to manage sales products so that they are optimized for AI open market rather than push-type product exposure through high-cost advertisement execution.

In the search-based AI open market, AI products that are preferred by AI systems must be sold in order to be exposed to product recommendations and searches. Therefore, there is a need for a solution that allows small business sellers to predict the sale of exposed products by exposing products to users at an applied cost for the initial start-up or advertising costs.

Korean Patent Laid-Open Patent No. 10-2018-0002944 (published on September 9, 2018)-Big data recommendation system and recommendation method for optimal product proposal for each individual preference

The present invention is to provide an AI keyword recommendation system and method based on open market big data that can search and optimize sellers' products in an artificial intelligence (AI)-based open market.

The present invention does not require SDK work for small business sellers who directly retail products, and can easily register and use products for sale, and monitor search optimization by providing search ranking trends of optimized products, and ranking trend insights. It is to provide an AI keyword recommendation system and method based on open market big data that enables product-specific inventory management prediction.

Objects other than the present invention will be easily understood through the following description.

According to an aspect of the present invention, as a system for recommending keywords based on artificial intelligence (AI) using open market big data, a big data collection engine that collects data on a customer's journey for customers shopping on an open market platform. ; A big data analysis module that analyzes data collected by the big data collection engine and performs keyword analysis for each seller product; And a customer channel module that recommends exposure-optimized keywords for each seller product based on the keyword analysis result. An AI keyword recommendation system based on open market big data is provided.

The big data analysis module includes: a data mining unit that analyzes the data through data mining according to at least one analysis algorithm among data cluster analysis (DCA), single linkage method (SLM), and association rule discovery (ARD); And a learning analysis unit that performs keyword analysis for each seller product through machine learning based on the analysis result of the data mining unit.

The big data analysis module further includes an image recognition unit for classifying images related to seller products, reading objects, and derive keywords for the objects, wherein the learning analysis unit also includes keywords derived from the image recognition unit in keyword analysis. Can be utilized.

The learning analysis unit collects tag keywords matched with the seller product as a reference meta tag, and derives at least one of a related search word, an auto-complete search word, and a shopping related search word as an extended keyword for the reference meta tag, or the seller product A product tag exposed higher in a product category to which it belongs may be included as a related keyword, or a keyword of a current product that is searched and exposed from a tag keyword of a related product may be included as a related product related keyword.

The customer channel module may include a keyword recommendation unit that tracks the exposure frequency and search ranking of seller products based on the keyword analysis result, derives new keywords not set in the seller products, and recommends them as the exposure optimization keywords. have.

The keyword recommendation unit may select a recommended keyword so that the seller product indicates a relatively high search ranking and automatically reflect the seller product.

Meanwhile, according to another aspect of the present invention, as an open market big data-based AI keyword recommendation method performed in an open market big data-based AI keyword recommendation system, (a) collecting keywords matched for seller products as a reference meta tag ; (b) performing a search in an open market for a keyword population in which one or more of the extended keyword, the related keyword, and the related product related keyword are combined with respect to the reference meta tag; (c) analyzing a search execution result to select a recommended keyword representing a high ranking of the seller product; And (d) recommending the recommended keyword to the seller product. An AI keyword recommendation method based on open market big data is provided.

Prior to the step (b), (a1) deriving at least one of a related search word, an autocomplete search word, and a shopping related search word as an extended keyword for the reference meta tag; (a2) collecting product tags exposed higher in a product category to which the seller product belongs and including them as related keywords; (a3) performing one or more of the steps of including the keyword of the current product, which is being searched and exposed in the tag keyword of the related product, as a related product related keyword, and the step (b) is the extended keyword and the It is possible to perform a search in the open market for a keyword population combining at least one of a related keyword and the related product related keywords.

In the step (d), a keyword indicating a high ranking of similar sales products of other sellers other than the seller product may be selected as the recommended keyword.

Other aspects, features, and advantages other than those described above will become apparent from the following drawings, claims, and detailed description of the invention.

According to an embodiment of the present invention, there is an effect of searching and optimizing a seller's product in an artificial intelligence (AI)-based open market.

In addition, for small business sellers who sell products directly, there is no need for SDK work, and you can easily use them by registering products in bulk, and you can monitor search optimization by providing the trend of search rankings of optimized products, and provide ranking trend insights. It also has the effect of making it possible to predict inventory management by product.

1 is a block diagram of an open market big data based AI keyword recommendation system according to an embodiment of the present invention;
2 is a block diagram of a big data collection engine;
3 is a flow chart of a method for recommending an AI keyword based on open market big data according to an embodiment of the present invention.
4 to 6 are diagrams each showing a seller product page, information on a higher-ranked search product, and an image recognition result screen.

In the present invention, various modifications may be made and various embodiments may be provided, and specific embodiments will be illustrated in the drawings and described in detail. However, this is not intended to limit the present invention to a specific embodiment, it is to be understood to include all changes, equivalents, or substitutes included in the spirit and scope of the present invention.

When a component is referred to as being "connected" or "connected" to another component, it is understood that it may be directly connected or connected to the other component, but other components may exist in the middle. Should be. On the other hand, when a component is referred to as being "directly connected" or "directly connected" to another component, it should be understood that there is no other component in the middle.

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

The terms used in the present specification are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as "comprise" or "have" are intended to designate the presence of features, numbers, steps, actions, components, parts, or a combination thereof described in the specification, but one or more other features. It is to be understood that the presence or addition of elements or numbers, steps, actions, components, parts, or combinations thereof, does not preclude in advance.

In addition, components of the embodiments described with reference to each drawing are not limited to the corresponding embodiments, and may be implemented to be included in other embodiments within the scope of maintaining the technical spirit of the present invention. Even if the description is omitted, it is natural that a plurality of embodiments may be implemented again as a unified embodiment.

In addition, in the description with reference to the accompanying drawings, the same or related reference numerals are assigned to the same components regardless of the reference numerals, and redundant descriptions thereof will be omitted. In describing the present invention, when it is determined that a detailed description of related known technologies may unnecessarily obscure the subject matter of the present invention, a detailed description thereof will be omitted.

In addition, terms such as "... unit", "... unit", "... module", and "... group" described in the specification mean a unit that processes at least one function or operation, which is hardware or software or hardware and software. It can be implemented by a combination of.

1 is a block diagram showing the configuration of an AI keyword recommendation system based on open market big data according to an embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of a big data collection engine, and FIG. It is a flow chart of a method for recommending an AI keyword based on open market big data, and FIGS. 4 to 6 are diagrams showing a seller product page, information on a higher-ranked search product, and an image recognition result screen, respectively.

The existing e-commerce solution field is focused on implementing a system that integrates sales product management, order management, and customer management. Since 3rd party e-commerce solution providers do not have big data of customer activity (search history, purchase pattern, etc.), efficiency of sales management is achieved with sales/order/customer information obtained through API officially provided by the e-commerce platform. , It was only possible to implement the integration function.

Market demand is changing into a field of sales marketing analysis solutions that can directly increase sales by activating sales in the field of solutions that efficiently store and manage big data collected from existing e-commerce platforms. Data marketing analysis technology that focuses on sales activation based on big data such as search and purchase patterns of customer activities collected by major e-commerce platform providers is growing.

In addition, the AI open market, a growing field of e-commerce, provides a big data-based sales marketing analysis function that allows platform providers to provide their own excellent sales management system and increase seller sales.

However, the AI open market is managed mainly by sellers with high sales, and at the same time, it is concentrated on the implementation of marketing functions on the consumer side, so sellers who are small businesses must market themselves in a sales method optimized for the sales channel platform technology.

The AI-based keyword recommendation system and method using open market big data according to an embodiment of the present invention is a seller product-based recommendation system, not a platform-based recommendation system, and serves a seller who wants to sell an arbitrary product on the open market. It is characterized in that it searches and optimizes seller products as users and recommends them to the platform. The open market is the main market sector, and it induces increased inflow to new customers due to product exposure. API/SDK development according to the e-commerce platform is unnecessary, and can be applied through product CSV upload.

Referring to FIG. 1, the open market big data based AI keyword recommendation system 100 according to the present embodiment includes a big data collection engine 110, a big data analysis module 120, and a customer channel module 130.

The big data collection engine 110 collects data on a customer movement path, such as customer purchase activity and search history, shopping on an open market platform. The big data collection engine 110 may periodically collect a purchase pattern history, related search product information (photos, product information, etc.), related search words, and the like among data based on a customer's shopping activity.

The big data collection engine 110 may include an Execute Main, a Foundation, a Process, and an Element (see FIG. 2).

The Excute Main part is the beginning of ESSD and is responsible for setting the environment for execution of the upper element, configuring the API required for the ESSD element, and running the process.

Excute Main part includes XML parser, DB pool manager, log controller, IPC manager, process controller, load environment module, etc. Main functions include initialization of environment variables, operation of process, operation of community server, operation of schedule, etc., and function of recording and saving the value of process result. Foundation is initialized by Excute Main and includes APIs required during operation of upper program such as Interface (IPC Community, Socket Server) between processes in upper area, Utility API in upper area, Log API to save detailed records of operation status, etc. do. The Foundation division includes the log module, IPC community module, utility module, and web drive loader.

Process is driven by Excute Main, and the number and driving type (independent program in one-time or Daemon type) are also determined by Excute Main. The Process part includes a One life cycle process module and a daemon process module. Process drives the upper element and delivers the result value to Excute Main. Selection of the type of operation and operation type of the element is transmitted from Excute Main, and communication is communicated through the IPC Community in the Foundation area.

Element includes Web Crawler engine, Web Data Parser, specific Web data reading and Web Input automation function. Element is driven by receiving information about Web URL, function type, and parser by process, and delivers the data corresponding to the result to the process area through the IPC community in the foundation area.

Element Web Crawler engine type is composed of Selenium + JSoup, and each library is included.

Since the data capacity that the big data collection engine 110 stores in real time or periodically is large, a large-capacity distributed data storage technology can be applied.

Referring back to FIG. 1, the big data analysis module 120 analyzes the big data collected by the big data collection engine 110.

The big data analysis module 120 may include a data mining unit 121, an image recognition unit 122, and a learning analysis unit 123.

The data mining unit 121 analyzes big data through data mining. One or more analysis algorithms among DCA, SLM, and ARD may be applied to the analysis of big data.

DCA (Data Cluster Analysis) allows data points to form clusters when there are unknown characteristics in the data collected by the big data collection engine 110, and group them based on the similarity of these data characteristics (clustering). ) Again, it is a technique to create grouping data based on distribution within the group.

The Single Linkage Method (SLM) is a method of grouping into large clusters with high similarity corresponding to the shortest distance between clusters between two data.

ARD (Association Rule Discovery) is a method of discovering related keywords in the relationship data between shopping customers and commerce products, and extracting recommended keywords that can optimize the exposure of seller products.

The image recognition unit 122 classifies images related to the product and reads the object. And it derives keywords of objects in the image. The image recognition unit 122 may be provided outside, for example, may utilize an image recognition API (image recognition API) according to computer vision provided by Google Cloud.

The learning analysis unit 123 analyzes the keywords derived from the data mining unit 121 and the image recognition unit 122 for each seller product registered by a user of the system through machine learning. In addition, in relation to the seller products registered by the user, the e-commerce platform (AI open market) tracks and analyzes the search rankings of customers' shopping search trends by keyword.

The customer channel module 130 recommends exposure optimization keywords for each seller product. If the exposure optimization keyword is reflected in the product name or product tag, the search ranking may increase and the inflow of visitors may increase after a certain period (1 to 3 days) depending on the degree of optimization of the product.

To this end, the customer channel module 130 may include a keyword recommendation unit 131.

The big data collection engine 110 may first collect keywords related to search for each category of open market products, popular trend keywords for each product, and search-related keywords for each trend keyword. In addition, a second collection may be performed based on keywords extracted from product images registered by users, product description keywords, and product tag keywords. It is possible to track the exposure frequency and ranking (search ranking) of seller products based on the keyword extracted by matching the keywords collected first and the keywords collected secondarily.

By quantifying the exposure frequency and search ranking based on the extracted keywords of the seller product on a certain basis, a keyword that the seller has not set for the seller product can be derived and provided to the seller as an optimized keyword for each product. In this case, it is also possible to provide a trend of product rankings (search rankings for each seller product keyword) exposed for each search keyword.

If the seller adds or edits the recommended impression optimization keyword in the product name, description, tag, etc., the searched impression keyword may increase and the ranking may increase. In this embodiment, the customer channel module 130 may automatically add or modify the recommended exposure optimization keyword to one or more of a product name, a detailed description, and a tag.

In addition, the customer channel module 130 may include a trend prediction unit 132. The trend prediction unit 132 may predict popularity by product category and recommend a user preferred product group (population prediction keyword trend by product category) by category in the open market. And when registering a new product, you can also recommend a product group (keyword) that is optimized for the open market in a short period of time. When registering a product with the featured keyword, the exposure rate can be significantly increased.

In addition, the trend prediction unit 132 may monitor/analyze the products of the shopping mall designated by the user as a competitor and provide a search trend for each product.

In addition, the customer channel module 130 may include a product management unit 133. The product management unit 133 may inform the service user of an optimization issue (ranking change, inflow change, etc.) report on the seller product registered by the user in a predetermined manner (SMS sending, mobile web page linking, etc.).

The product management unit 133 may provide inventory management and order management for seller products. Through the order status dashboard, it is possible to provide product-waiting status for each customer and order statistics for each product.

In addition, the product management unit 133 may manage the lowest price for the price comparison product. It can support price comparison ranking, release status, and lowest price automation within products bundled with price comparison.

Referring to FIG. 3, a method of recommending AI keywords based on open market big data will be described. The method may be performed by the learning analysis unit 123 and the keyword recommendation unit 131.

In order to use an AI keyword recommendation solution based on open market big data, keywords (tag keywords) matched by customer-registered service users for seller products are collected (step S200). In addition, a combination keyword combining a product name and a category may be included as a reference meta tag.

For each reference meta tag, an extended keyword may be derived using a related search word, an auto-complete search word, and a shopping related search word provided by the portal site and/or the shopping mall site.

Product tags that are exposed to the top among product name/tag keywords for each product category and search product name/tag keywords for each keyword, that is, frequently used, may be collected and included as related keywords.

In addition, the keyword of the current product that is searched and exposed from the tag keyword of the related product may be included as a related product related keyword.

Filtering such as removing keywords that are not related to the category of seller products or removing duplicates may be performed on these various keywords.

The result of searching in the open market is analyzed using the keyword population including the collected keywords (step S210).

As a result of the analysis, recommended keywords indicating a high ranking of seller products are selected and recommended to service users (step S220).

When selecting a keyword in step S220, a keyword indicating a high ranking of similar sales products of other sellers other than the sales product of the corresponding seller may be selected and recommended to the corresponding seller. The criteria for calculating recommended keywords are as follows. Each numerical value below is exemplary and may be changed as necessary.

A maximum of 30% can be given depending on whether or not the search exposure is based on a 3% addition standard for each of 10 keywords. Add 3% for each keyword that ranks out of 10 tag keywords. If all 10 are exposed, it could be 30%.

A maximum of 30% can be given according to the ranking based on the 3% addition of 10 keywords. Among the exposed keywords, if the ranking is within the designated ranking range (eg, 1, 2P), it is considered to be applicable. You can add 3% per keyword. And you can add up to 20% depending on the number of impressions and clicks. If the number of 100-200 clicks (inflow) per product is 10%, and if it is more than 201 clicks (inflow), 15% is added. If the number of impressions per product is more than 5000, it is added as 5%, and if it is less than 5000 impressions, it is added as 0%

And, depending on the number of orders, it can be added within the range of 20%. If the number of orders per product is 10-20, 10% is given, 0% for less than 10 items, and 20% for more than 21 items.

If all of the above conditions are satisfied, it becomes 100%, and the higher the %, the higher the keyword optimization rate for each product. The numerical standard of each item is daily data, and may be data from the previous day.

The keyword recommendation algorithm will be described with reference to drawings and examples for convenience of understanding and explanation of the invention.

Currently, the following categories, product names, and tags may be set for the seller product (see FIG. 4).

Category: Fashion Clothing> Women's Clothing> Coat

Product Name: [SALE] Winter Slim Women's Wool Code (Same Day Delivery)

Tags: #wool code #winter girl coat #female coke #girl coat #girl wool coat #girl slim coat #winter coat #shorter girl coat #woman winter coat

In this case, in step 1, check whether the product name has the same character as the category name. Only the product name is used as the default tag (keyword) only if there is a product name, and if not, the product name + category name is used as the default tag.

In step 2, the portal site (e.g. Naver) searches for auto-complete, related search terms, and related search terms in representative categories.

In step 3, the top 5 tag information of the product list output from each product representative category is collected using the search word and product name. In addition, the top 5 tag information of the product list output from each product detail category is collected using the search word and product name. In this case, the top five products can be selected as having item tag information (see FIG. 5).

In step 4, image analysis information is collected through the image recognition unit 122 using the registered product image or through an API of an external image recognition device (eg, Google) (see FIG. 6).

In step 5, the same tag (keyword) is removed.

In step 6, tags are extracted from the conditional search results for each keyword. In this case, you can use the API provided by the open market. For example, in the case of Naver Shopping API, it is the same as https://developers.naver.com/docs/search/shopping/.

Using the final tag (keyword) finally derived in step 7, the priority of the keyword for the product in the open market is determined.

In the present embodiment, the open market big data-based AI keyword recommendation method may be implemented in the form of a recording medium including instructions executable by a computer such as an application or program module executed by a computer. Computer-readable media can be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Further, the computer-readable medium may include a computer storage medium. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

The above-described method can be executed by an application basically installed on the terminal (which may include a program included in the platform or operating system basically mounted on the terminal), and the user can It may be executed by an application (ie, a program) directly installed on the master terminal through an application providing server such as a web server. In this sense, the above-described method may be implemented as an application (ie, a program) installed basically in a terminal or directly installed by a user, and recorded in a computer-readable recording medium such as a terminal.

Although the above has been described with reference to embodiments of the present invention, those of ordinary skill in the art can variously modify the present invention within the scope not departing from the spirit and scope of the present invention described in the following claims. And it will be appreciated that it can be changed.

100: AI keyword recommendation system based on open market big data
110: big data collection engine 120: big data analysis module
121: data mining unit 122: image recognition unit
123: learning analysis unit 130: customer channel module
131: keyword recommendation unit 132: trend prediction unit
133: product management unit

Claims (9)

As a system that recommends keywords based on artificial intelligence (AI) using open market big data,
A big data collection engine that collects data on a customer's travel path for customers shopping on an open market platform;
A big data analysis module that analyzes data collected by the big data collection engine and performs keyword analysis for each seller product; And
AI keyword recommendation system based on open market big data including a customer channel module that recommends exposure-optimized keywords for each seller product based on the keyword analysis result.
The method of claim 1,
The big data analysis module,
A data mining unit for analyzing the data through data mining according to at least one analysis algorithm among data cluster analysis (DCA), single linkage method (SLM), and association rule discovery (ARD); And
AI keyword recommendation system based on open market big data comprising a learning analysis unit that performs keyword analysis for each seller product through machine learning based on the analysis result of the data mining unit.
The method of claim 2,
The big data analysis module further includes an image recognition unit for classifying images related to seller products, reading objects, and derive keywords for the objects,
An AI keyword recommendation system based on open market big data, characterized in that the learning analysis unit also utilizes the keyword derived from the image recognition unit for keyword analysis.
The method of claim 3,
The learning analysis unit collects tag keywords matched for the seller product as a reference meta tag,
For the reference meta tag, at least one of a related search word, an auto-complete search word, and a shopping related search word is derived as an extended keyword,
Includes a product tag that is exposed higher in the product category to which the seller product belongs as a related keyword, or
AI keyword recommendation system based on open market big data, characterized in that the keyword of the current product, which is being searched and exposed from the tag keyword of the related product, is included as a related product related keyword.
The method of claim 1,
The customer channel module tracks the exposure frequency and search ranking of seller products based on the keyword analysis result, and includes a keyword recommendation unit that derives a new keyword not set in the seller product and recommends it as the exposure optimization keyword. AI keyword recommendation system based on market big data.
The method of claim 5,
The keyword recommendation unit selects a recommended keyword so that the seller product indicates a relatively high search ranking, and automatically reflects the recommended keyword on the seller product.
As an open market big data-based AI keyword recommendation method performed in an open market big data-based AI keyword recommendation system,
(a) collecting keywords matched for seller products as a reference meta tag;
(b) performing a search in an open market for a keyword population combining at least one of the extended keyword, the related keyword, and the related product related keyword with respect to the reference meta tag;
(c) analyzing a search execution result to select a recommended keyword representing a high ranking of the seller product; And
(d) an AI keyword recommendation method based on open market big data comprising the step of recommending the recommended keyword to the seller product.
The method of claim 7,
Prior to step (b),
(a1) deriving at least one of a related search word, an autocomplete search word, and a shopping related search word as an extended keyword for the reference meta tag;
(a2) collecting product tags that are exposed higher in the product category to which the seller product belongs and including them as related keywords;
(a3) including a keyword of a current product that is searched and exposed in a tag keyword of a related product as a related product related keyword;
Do one or more of,
The step (b) is based on open market big data, characterized in that the search is performed in an open market for a keyword population that combines at least one of the extended keyword, the related keyword, and the related product related keyword with respect to the reference meta tag. How to recommend AI keywords.
The method of claim 7,
The step (d) is an open market big data based AI keyword recommendation method, characterized in that the keyword representing a high ranking of similar sales products of other sellers, not the seller product, is selected as the recommended keyword.

Images