KR102610665B1 - Main product recommendation server and method for the user without login using category recommendation - Google Patents

Abstract

The non-logged-in user product recommendation server of the present invention receives usage information from the user terminal when a non-logged-in user accesses the mobile open market and performs one or more of product inquiry, wish list, shopping cart, and purchase, and generates recommendations. A transmitter and receiver that transmits category product information to the user terminal, saves the latest usage information in the form of time information, calculates a category interest score based on this, and inputs the calculated category interest score into the ALS algorithm to calculate category score information. It is characterized by including a control unit that generates personalized recommended category product information by combining the calculated category score information and category information of the mobile open market main product. According to the present invention, since recommended category product information is generated based on usage information for each user, there is an effect of recommending a personalized main product by transmitting the recommended category product information to the user terminal.

Description

Non-logged user main product recommendation server and method using category recommendation {MAIN PRODUCT RECOMMENDATION SERVER AND METHOD FOR THE USER WITHOUT LOGIN USING CATEGORY RECOMMENDATION}

The present invention relates to a mobile open market product recommendation technology. More specifically, the present invention relates to a mobile open market product recommendation technology, and more specifically, uses the latest usage information of non-logged-in users such as product inquiry, wish list, shopping cart addition, and purchase in the mobile open market to reflect recency, and ALS (Alternating Least Square) algorithm is used to generate recommended category product information based on the latest usage information, and when accessing the mobile open market screen, recommended category main product information can be displayed at the top according to non-logged user preferences. It relates to recommended servers and methods.

Recently, a lot of e-commerce is taking place through open markets through mobile devices. Individuals or small businesses selling products can benefit from selling products at low prices without intermediate distribution margins, and consumers can conveniently purchase the products they want online. This is because it has the advantage of being purchased at a low price, so it can kill two birds with one stone. However, there are problems with trading through mobile open markets.

First, let's look at the problems of transactions through mobile open markets. Most open markets stimulate the desire to purchase by displaying recommended products according to the logged in user's preferences on the user's terminal screen when accessing the mobile screen. In this case, the user must log in to analyze information about the user and recommend products in the category of interest. However, if recommended products are provided to only logged-in users like this, it is not possible to analyze the categories of interest for non-logged-in users and products cannot be recommended, so a potential increase in sales cannot be expected. Additionally, when a non-logged-in user accesses the mobile open market screen, the open market recommends products through various methods. When information is analyzed based on usage information over a certain period of time rather than the user's most recent usage information, past interest levels are also included. As a result, the user's latest interest level cannot be accurately reflected. As a result, non-logged-in users receive recommendations for products they are not interested in, which creates a problem in that they do not lead to the purchase of the recommended products.

To solve this problem of product recommendation for non-logged-in users, product recommendation for non-logged-in users using collaborative filtering has been proposed. This also has another problem. Recent open markets use User Based Recommend or Item Based Recommend during collaborative filtering to provide recommended product information to users. At this time, when the user performs any one of viewing, wishing, adding to shopping cart, and purchasing the recommended product, There is a problem in that the corresponding product may be excluded from the next recommendation task and recommendations may be made for products that the user is not interested in.

Product recommendation for non-logged users also has its problems, but the non-logged method will continue to be used due to the advantages of users being able to use the service without the inconvenience of logging in and receiving recommendations for products they are interested in from the open market. It is expected that In this case, a function that calculates interest scores for each category by analyzing usage information such as product inquiry, wish list, shopping cart addition, and purchase performed while the user is not logged in, and the user can view recommended products, wish list, and shopping cart addition. A function to generate recommended category product information based on the category score calculated using the ALS algorithm that does not exclude the product from the next recommendation task when one of the following is performed, and a personalized main product based on the recommended category product information A function to recommend is required.

Republic of Korea Patent Publication No. 10-2015-0101340 (2015.09.03)

The present invention calculates a category interest score based on usage information such as product inquiry, wishlisting, adding to shopping cart, and purchase that a non-logged-in user has recently performed in a mobile open market through a user terminal, and calculates a category score based on the category interest score. The purpose is to provide a product recommendation server and method that generates recommended category product information based on category scores after calculating and transmits the generated information to the user terminal to recommend the main product of the recommended category to non-logged-in users.

In addition, the purpose is to provide a product recommendation server and method that applies the number of product clicks as a weight to the latest usage information of non-logged-in users when calculating the category interest score and reflects it in the recommended product ranking.

In addition, when a non-logged-in user performs any of the following: viewing, adding to a wish list, adding to cart, or purchasing a recommended product, recommended category product information based on interest score calculated using the ALS algorithm that does not exclude the product from the next recommendation task The purpose is to provide a product recommendation server and method for generating.

Additionally, the purpose is to provide a product recommendation server and method that recommends personalized main products based on recommended category product information.

Meanwhile, the technical problem to be achieved by the present invention is not limited to the technical problems mentioned above, and various technical problems can be derived from the contents described below within the range obvious to those skilled in the art.

The product recommendation server according to the present invention for achieving the above purpose includes a transmitter and receiver that receives usage information of non-logged-in users from a user terminal and transmits the recommended category product information to the user terminal, and provides the latest information based on the usage information. A control unit that calculates a category interest score reflecting the last name, calculates a category score using the category interest score, and then combines the category score information with the category information of the mobile open market main product to generate the recommended category product information, It is characterized by comprising a storage unit storing usage-related information related to usage information, the category interest score, the category score, category information related to the mobile open market main product, and the recommended category product information.

Additionally, the usage information may be stored as the latest usage point information for non-logged-in users.

Additionally, the usage information may include mid (Member id) stored in a cookie when the user logs in at least once.

Additionally, the usage information may include one or more of the major, medium, and small categories in product classification.

Additionally, in the process of calculating the category interest score, the number of product clicks may be applied as a weight to the latest usage information of non-logged-in users.

Additionally, the category score can be calculated from an ALS algorithm that does not exclude the product from the next recommendation task when any of the recommended products are viewed, added to a wish list, added to a shopping cart, or purchased by a non-logged-in user.

Meanwhile, a product recommendation method according to another embodiment of the present invention includes (a) receiving usage information from a user terminal, (b) storing the usage information, and (c) updating the latest information based on the usage information. Calculating a category interest score reflecting the (d) calculating a category score using the category interest score, (e) combining the category score information with the category information of the mobile open market main product to provide recommended category product information It is possible to provide a product recommendation method including the step of generating, and (f) transmitting the recommended category product information to the user terminal when a non-logged-in user accesses the mobile open market.

According to the present invention, when a non-logged-in user accesses the mobile open market through a user terminal, the product recommendation server stores usage information such as the user's product inquiry, wish list, addition to shopping cart, and purchase, and creates an interest score for each category based on this. After calculating the category score using the category interest score, the recommended category product information generated by combining the category score and the category information of the mobile open market main product is transmitted to the user terminal to recommend a personalized main product. There is an effect that it can be done.

In addition, since the category interest score is calculated based on the latest usage information of non-logged-in users, recency can be reflected. At this time, when calculating the category interest score, the number of product clicks can be applied as a weight and reflected in the recommended product ranking. When a non-logged-in user performs any of the following: viewing, adding to a wishlist, adding to cart, or purchasing a recommended product, ALS does not exclude that product from the next recommendation task and can continuously recommend product information of interest to the non-logged-in user. Using an algorithm, it is effective in generating and providing recommended category product information that non-logged-in users are interested in.

The effects of the present invention are not limited to the effects mentioned above, and various effects may be included within the scope apparent to those skilled in the art from the contents described below.

1 is a diagram showing the overall configuration of a product recommendation server according to an embodiment of the present invention.
Figure 2 is a diagram showing how product information according to category classification is output to a user terminal.
Figure 3 is a diagram showing how detailed information about the product that can be confirmed when a category and product of interest is selected is output to the user terminal.
Figure 4 is a diagram showing a flow chart between the transmitting and receiving unit, the control unit, and the storage unit according to the present invention.
Figure 5 is a diagram showing a flow chart between a user terminal and a product recommendation server according to the present invention.
Figure 6 is a diagram showing the connection structure between a user terminal and a product recommendation server according to the present invention.
Figure 7 is a diagram showing the result of transmission of recommended category product information from the transmitter and receiver to the user terminal.

Hereinafter, some embodiments of the present invention will be described in detail through illustrative drawings. The described embodiments are provided so that those skilled in the art can easily understand the technical idea of the present invention, and the present invention is not limited thereto, and it is judged that detailed description of related known configurations or functions may obscure the gist of the present invention. In this case, the detailed description is omitted.

In addition, the matters expressed in the attached drawings may be different from the form actually implemented in the schematic drawings to easily explain the embodiments of the present invention, and in adding reference numerals to the components in each drawing, the same configuration It should be noted that elements, even if shown in different drawings, are given the same symbols as much as possible.

Additionally, the expression 'including' certain components is an 'open expression' that simply refers to the existence of the components, and should not be understood as excluding additional components.

Figure 1 is a diagram showing the overall configuration of the product recommendation server 100 according to an embodiment of the present invention.

Referring to FIG. 1, the product recommendation server 100 includes a transceiver 110, a control unit 120, and a storage unit 130, and other additional components necessary for configuring the device may also be included. Of course.

Meanwhile, with regard to the product recommendation server 100, there is a need to first explain the server. The computer hardware on which a server program is running is called a server, and a computer program that provides services to other programs is also called a server. The server monitors and controls the entire network, such as printer control or file management, or connects to other networks through mainframes or public networks, and manages software resources such as data, programs, and files, or hardware resources such as modems, faxes, shared printers, and other equipment. It plays a role in helping to share.

In addition, the server provides services at the request of the user (client), and a system configured in this way is called a client/server system, which refers to a form of distributed processing in which one or more application programs are operated in a cooperative environment. In other words, the client/server system is a processing method that obtains the desired results desired by the user through cooperation between the client requesting the service and the server processing the client's request. In the case of a small LAN with about 5 to 20 clients, one device can sufficiently handle all services, but in the case of a large LAN, multiple devices are deployed, and file management is performed by a file server and printer control is performed by a printer server. Each role is subdivided, with the communication server taking charge of external exchanges such as the Internet. An online server is a computer that discloses information contained in its hard disk to the outside world in a special form. Generally, online, various information is managed on a server, and general users use their computers (clients) to access the server and use the information provided by the server. Accordingly, the product recommendation server 100 can be configured as an independent server that only has a product recommendation function or product information, or it can also be implemented as a component of another device with a product recommendation function or product information. Hereinafter, the components included in the product recommendation server 100 will be described.

The transceiver unit 110 receives usage information from the user terminal 200 when a non-logged-in user accesses the mobile open market through the user terminal 200 and performs any one or more of product inquiry, wishlisting, addition to shopping cart, and purchase. Receives and delivers recommended category product information generated based on non-logged-in user usage information to the user terminal 200. The user terminal 200 includes a smart phone, a tablet computer, and information and communication devices. If the user terminal 200 is a smartphone, a non-logged-in user may download and install the device on the user terminal 200. You can enter information or check printed product information through the application.

The control unit 120 calculates a category interest score based on non-logged-in user usage information stored in the storage unit 130. Here, the calculated category interest score applies the number of product clicks as a weight to the information on the time of use by non-logged-in users for the time of product view, time of wishing, time of adding to shopping cart, and time of purchase for each major, medium, and small category. is calculated. At this time, the interest score is calculated by applying a greater weight to the time of adding to the shopping cart than to the time of wishing and viewing the product, and additionally applying the number of product clicks as a weight. This is because the number of product clicks is applied as a weighting factor. This is because the probability of purchasing a product is higher if the product has been added to the shopping cart more recently, and the number of clicks on the product increases as the time of purchase approaches.

Thereafter, the control unit 120 analyzes the information by transmitting the category interest score as an input value of the ALS algorithm among the collaborative filtering types of User Based Recommend, Item Based Recommend, and ALS algorithms. The input values of the ALS algorithm are the user identifier mid, category id, and the category interest score. At this time, the category id is a unique value designated to distinguish each major category, medium category, and small category. For example, for shoes, which are major categories, sneakers as a subcategory of shoes, and running shoes, which are subcategories as a subcategory of sneakers, which are middle category categories, can be specified as 1 as a unique value of the category id, and for home appliances, which are major categories, lifestyle. For telephones, which are a mid-category category as a subcategory of home appliances, and landline phones, which are a small category as a subcategory of telephones, 20 can be specified as a unique value of the category id. In addition, 44 can be specified as a unique value of the category id for fresh food, which is a major category, and domestic fruits, which are a subcategory of fresh foods, and citrus, which is a small category as a subcategory of domestic fruits, which is a medium category. In addition, the category ID can be assigned as a unique value to several other products according to each category classification in the same manner as above, and an example of the category ID is shown in Table 1 below.

category id Main Category
category id Middle classification
category id Subcategory
category id One One 34 22 20 2 32 24 44 3 31 25 55 One 55 26 60 7 35 111

Additionally, Lambda, Feature Number, and Iterations can be specified as input values for the ALS algorithm. Lambda, Feature Number, and Iterations are values used for tuning, so they can be changed and used to suit the purpose of use.

Meanwhile, the reason for not using User Based Recommend or Item Based Recommend is that the product recommendation server 100 recommends the main product to non-logged-in users, and non-logged-in users cannot view, add to wishlist, add to shopping cart, or purchase the recommended product. If either one is performed, the corresponding product is excluded from the next recommendation task, and the next recommendation may recommend products that non-logged-in users are not interested in.

Meanwhile, there is a need to explain collaborative filtering. Collaborative filtering is a type of recommendation algorithm that determines a user recommendation group of people with similar preferences to the user and reflects the opinions of the determined user recommendation group to predict the preference of products that the user has not yet purchased and products that are predicted to have high preference. This is a recommended technique.

A recommendation technique based on conventional collaborative filtering includes the steps of determining people's preferences for specific products, determining a user recommendation group with similar preferences to the user based on the determined preferences, and determining the user's preference in the determined user recommendation group. It consisted of recommending products that were expected to have high prices.

First, looking at the step of determining people's preferences in more detail, in the conventional collaborative filtering-based recommendation technique, the input data consists of a set of preferences of m customers for n products, which is a customer-product matrix (P) of size n?m. It is expressed as Preferences for specific products are usually entered directly from customers or implicitly estimated from purchase data. For example, when measuring preference from purchase data, the value pij of the ith row and jth column of the customer-product matrix (P) has a value of 1 if the ith customer purchased the jth product, and 0 otherwise. Each has a value.

Next, looking more specifically at the step of determining a user recommendation group with similar preferences to the user, the step of determining the user recommendation group is to calculate the degree of preference similarity between the user and other customers using the customer-product matrix (P). do. In other words, it is a process of searching for i user recommendation groups with the most similar preferences for each customer u. In general, Pearson Correlation is used to measure the degree of similarity between users and other customers.

The Pearson correlation coefficient is a correlation coefficient used to determine the relationship between two variables. For example, when both variables to be analyzed, such as systolic blood pressure and diastolic blood pressure, or height and weight, are continuous data, it is a linear correlation between two variables. It is a value that represents the size in a parametric way.

Meanwhile, looking at the product recommendation step in more detail, N recommended products that the user is expected to purchase or prefer are determined based on products purchased or preferred by customers constituting the determined user recommendation group. Most-frequent item recommendation is generally used as a standard for determining recommended products. Frequently purchased product recommendation is a method of analyzing the purchase history data of the user recommendation group and determining the top N products with high purchase frequency as recommended products.

When the ALS algorithm of collaborative filtering is performed, Feature and Rating are calculated as output values, and finally Matrix Factorization is performed using the output Feature and Rating values. At this time, the maximum rating value is set to 1 so that the rating ranges from 0 to 1. Since many recommended category results are generated when using the matrix factorization operation, filtering is performed only up to the top 100 by considering the rating value.

Meanwhile, the category score results generated through the ALS algorithm are stored in Memcached. Here, Memcached is a memory-based caching system that supports high performance. In general, it is used to increase the performance of the web system by loading data stored in the database into memory to increase response speed. Unless the website provides fixed content, it must read data from various storage devices called databases according to the user's request and respond to the user through a web server. If there are not many users, it does not matter, but if the number of users increases, a method must be prepared to increase the response speed to user requests. In this case, one of several methods to choose is to provide a cache server using memory. When there is a user request, the cache memory is checked to see if the corresponding information about the request exists, and if so, the corresponding information is read from the cache memory and provided to the user. If it is not in the cache memory, data stored on the disk, such as a database, is read, provided to the user, and stored again in the cache memory. The main functions of Memcached are reading, writing, and deleting. The advantage is that it has a fast response speed because it reads data from cache memory rather than a database, and it has excellent scalability because it can process data by distributing it across multiple servers. Disadvantages are that because data is stored in cache memory and provided when needed, data is not maintained when rebooting, and due to the key-value data structure, it does not provide LIKE search provided by relational databases.

In computer science, a cache refers to a temporary location where data or values are copied in advance. This is used when it takes longer to access the original data compared to the cache access time or when you want to save time recalculating the value. By copying data to the cache in advance, you can access the data at a faster rate without computation or access time. Types of cache include CPU cache and disk cache. CPU cache is a small memory mounted inside or right next to the CPU chip to speed up access to large amounts of main memory. Since the 1980s, most CPUs have been equipped with one or as many as five to six CPU caches with different functions. Because CPU processing speeds are increasing much faster than memory access speeds, small-capacity but high-speed CPU caches have a direct impact on the performance of modern microprocessors. Disk cache is also called disk buffer. Most hard disks sold since the late 1980s have a small computer built in for disk control and external interface. These small computers usually have a small memory that stores input/output data to the disk, which is called a disk cache. Disk cache is used for several purposes, such as read-ahead, throttling, and write acceleration. Regarding read-ahead, when the operating system requests to read/write to the disk, the disk arm can read data only after a certain period of time has passed after placing the head on the corresponding track. However, since the operating system often requests subsequent sectors in addition to the first sector, the disk reads later data in advance and stores it in the cache to save repetitive access time to consecutive locations. Regarding speed control, disk input/output interfaces are usually much slower than the speed at which disk platters read data. Therefore, in order for the disk arm to read and write quickly, the data is briefly stored in the cache before being passed on to the interface. Regarding write acceleration, when the operating system requests to write to the disk, the computer inside the disk stores the data in the cache and can report to the operating system that the writing has finished before the disk arm has finished writing. This allows the operating system to quickly perform the next operation without having to wait for the disk write to finish. However, with this method, there is a risk that data may be lost or the file system may be damaged if the power is cut off before the disk arm completes writing.

Thereafter, the control unit 120 generates recommended category product information by combining the category score information with the category information of the mobile open market main product. For example, if men's cotton pants are calculated with high category score information, after combining them with the category information of the mobile open market main product, jeans or dress pants, which are in the same category as the cotton pants, can be recommended as the main product in the user's recommended category. there is.

The storage unit 130 stores usage-related information related to the latest usage information of non-logged-in users, such as product inquiry, wish list, shopping cart addition, and purchase, category interest score, category score, category information related to the mobile open market main product, and recommended categories. Save product information.

Use by a non-logged-in user refers to one or more of the following: product inquiry, wishlisting, addition to shopping cart, and purchase performed by a non-logged-in user by accessing the mobile open market. After accessing the mobile open market, non-logged-in users can select the category of the product they are recently interested in or want to purchase and perform one or more of the following: product search, wish list, add to cart, and purchase, or use recommendations provided on the mobile open market home screen. By clicking on the main product, you can perform one or more of the following: product inquiry, wish list, add to cart, and purchase.

There are various product categories, for example, a clothing category that sells women's clothing, men's clothing, children's clothing, sports clothing and underwear, a miscellaneous goods category that sells shoes, bags and accessories, large home appliances, household appliances and The main categories that users can select from are the home appliance category that sells kitchen appliances, the food category that sells fresh foods, processed foods, and health foods, and the leisure category that sells hiking, camping, golf, bicycle, fitness, and ball game supplies. It can be. Additionally, the middle category can be selected as a subcategory of the major category. For example, as a subcategory of women's clothing, top types of t-shirts, blouses and knits, bottom types of pants, jeans and skirts, outerwear types of coats, cardigans and vests, and as subcategories of sports clothing, men's sportswear and women's sportswear. , men's training clothes and women's training clothes, running shoes, sneakers, sneakers and slippers as a type of footwear, sports bags, sports hats, sports glasses, sports watches and sports goods as a type of sports miscellaneous goods, TVs as a large home appliance as a subcategory of home appliances, You can choose between a refrigerator and washing machine or household appliances such as a vacuum cleaner, iron, telephone, sewing machine, electric razor, and hair dryer. In addition, depending on the product, you can select a subcategory as a subcategory of the midcategory in the same way. After selecting a category, perform one or more of the following: product search, wishlist, add to cart, and purchase for the product you are interested in or want to purchase. can do. At this time, the order of use for non-logged-in users may vary depending on the situation. For example, before searching for a product in a category you are interested in, you can add it to a wish list without clicking on the product. You can later add the wish list products to your shopping cart and purchase them, and then search for the product. After obtaining the detailed information of the selected product, you can also perform one or more of the following: add to wishlist, add to cart, and purchase. Additionally, you can search for products to find detailed information about them, add them to your shopping cart, and then proceed with purchasing the products in your shopping cart.

Figure 2 is a diagram showing product information according to category classification.

Categories can be classified as described in the storage unit 130, and the category to which the product belongs can be selected to find a product of interest or purchase.

Figure 3 is a diagram showing detailed information on the product that can be confirmed when the category and product of interest are selected.

Examples of user usage information that the storage unit 130 collects in the database when a non-logged-in user selects a category and product of interest and performs one or more of product inquiry, wish list, shopping cart, and purchase are shown in Table 2 below. It's the same.

mid Main Category
category id Middle classification
category id Subcategory
category id Saved Views When to add to cart Product inquiry time Number of product clicks u1 One 34 22 2014.04.11
17:50:55 - 2014.04.11
18:50:50 10 u1 2 32 24 - 2014.04.10
17:30:40 2014.04.12
10:00:00 5 u1 3 31 25 - - 2014.08.11
17:50:55 5 u1 One 55 26 2014.02.13
12:50:15 2014.02.14
17:50:50 2014.02.16
13:50:30 15 u2 One 35 111 2013.04.10
17:50:55 2013.04.08
13:50:22 2013.04.12
17:50:50 20 u2 2 44 112 - - 2014.08.27
11:50:55 10

Referring to Table 2, the storage unit 130 stores non-logged-in user usage information, such as the time of wish list, the time of adding to the shopping cart, the time of product inquiry, and the number of product clicks, for each major category, medium category, and small category for each user identifier, mid. Save it to

At this time, ‘use’ means that the user receives a specific service through the mobile open market. For example, 'use' may mean that a user 'searches' for a specific product. For example, ‘use’ may mean that a user ‘purchases’ a specific product through an open market. Additionally, the number of clicks on a product may be the number of times a user views the product in an open market.

Meanwhile, the user identifier mid is information needed to recommend a personalized main product. It is stored and managed in a cookie when a non-logged-in user logs in to the mobile open market at least once, so it can be performed by a non-logged-in user even if the user is not logged in. Information such as product inquiry, wish list, shopping cart addition, and purchase is recognized through the user identifier mid. Here, a cookie is a type of hypertext record and refers to a small record information file that is installed on the Internet user's computer by the server used by the Internet user when he or she visits a website. The name cookie is taken from the Grimm fairy tale 'Hansel and Gretel', and comes from the story that Hantel and Gretel dropped cookie pieces to mark the path they took. The information contained in this record file is read each time an Internet user visits the same website and is frequently replaced with new information. This method was invented by Lou Montulli, a program developer at Netscape, and is used by many servers and websites today to improve browser speed.

Hypertext is text that gives users immediate access from one document to another through references. It is mainly displayed through computers and other electronic devices, and in combination with the Internet, it has become the main component of HTML. While existing documents have a sequential and sequential structure, hypertext has a random and serial structure in which the order changes depending on the link. In other words, documents linked by reference can be freely moved according to any action (Click).

Meanwhile, a database refers to a set of data that is integrated and managed for the purpose of being shared and used by multiple people. Also known as DB (DataBase) for short, it is the heart of an information service organization that provides necessary information to a specific number of users or systematically accumulates information needed within an organization to provide necessary information to users within the organization. . Compared to the conventional method of processing files independently for each task, there are advantages such as no need to prepare new files even when the task expands, and the same data is not duplicated in each file. The system that manages the database is called a database management system. The characteristics of the database include real-time accessibility that immediately processes and responds to user inquiries, and dynamically displays the latest data through insertion, deletion, and update. , and multiple users can share desired data at the same time. When referring to data stored in a database, not only is it referenced according to the data content requested by the user rather than the table row address or location, but the application program and the database are made independent, so even if the logical structure of the data is changed, the application program does not change. Types of databases include hierarchical databases, network databases, relational databases, and object-oriented databases. Hierarchical databases use a hierarchical data model based on a tree structure. In the hierarchical data model, data is organized in the form of a tree, and each data element is composed of links that represent hierarchical relationships. Network-type databases use a network-type data model based on a graph structure. A network data model is a data model that connects entities and their relationships in a graph structure. This is similar to a hierarchical data model, but the difference is that it can have multiple parents (parent objects). Relational databases use a relational data model. Relational databases are widely used as they are known to be the most stable and efficient databases to date. The relational data model is an independent type of data model that uses objects as tables and connects them to each other using common properties between objects. Object-oriented databases are databases that appeared in the late 1980s and are based on object-oriented programming concepts. The object-oriented model has the advantage of organizing data and programs in the form of independent objects, making it easy to process complex data types and making objects easy to understand.

A database management system is a program that allows multiple computer users to record or access data in a database. The purpose is to provide a convenient and efficient environment for searching information in a database or storing information in a database. The database management system integrates scattered data for each application software and shares the integrated data with each application software to enable systematic use of information. The function of the database management system is to define the accumulated data structure, accumulate data according to the data structure, and search and update data using the database language. In addition, it is possible to execute data processing requests from multiple users simultaneously, and when an error occurs during update, it is possible to return to the state before update. The general type of database management system is a relational database management system, and the standardized user and program interface of the relational database management system is called SQL (Structured Query Language).

FIG. 4 is a diagram illustrating a flowchart between the transmitting and receiving unit 110, the control unit 120, and the storage unit 130 according to an embodiment of the present invention.

Referring to FIG. 4, when a non-logged-in user accesses the mobile open market and performs one or more of product inquiry, favorite, add to shopping cart, and purchase, the non-logged-in user's latest usage information is collected in the storage unit 130, The collected usage information is transmitted to the control unit 120 to calculate a category interest score.

An example of the category interest score calculated by the control unit 120 based on the latest usage information of non-logged-in users is shown in Table 3 below.

mid Main Category
category id Middle classification category id Subcategory category id category
Interest score u1 One 34 22 2.53 u1 2 32 24 0.76 u1 3 31 25 0.06 u2 2 44 112 31.3 u2 2 62 115 3.855

Referring to Table 3, the category interest score for each user can be calculated based on the latest point-of-use information for product viewing, wishlisting, adding to cart, and purchase performed by non-logged-in users and the number of product clicks applied as weight. At this time, the score is calculated by giving more weight to adding to cart than adding to wishlist and viewing products. This is because adding to cart has the highest probability of purchase, while adding to wishlist and viewing product results in a purchase with a lower probability than adding to cart. Because you lose. And the reason why the number of clicks on a product is applied as a weight is because the number of clicks on that product greatly increases as the time of purchasing the product gets closer.

The more recent the user's time of use for each category, the higher the category interest score, and the information that can be analyzed by calculating the category interest score can be used to determine which products in which category each user has recently been interested in.

Afterwards, an example of the category score calculated using the ALS algorithm based on the category interest score is shown in Table 4.

mid Main Category
Categoryid Middle classification category id Subcategory category id Rating u1 One 32 24 0.7 u1 2 89 - 0.4 u1 One 37 24 0.3 u2 One - - 0.3 u2 One 37 24 0.5

Referring to Table 4, category scores for each user can be calculated using the ALS algorithm of collaborative filtering. At this time, mid, category id, and category interest score are input to the ALS algorithm, and additional values such as Lambda, Feature number, and Iterations can be input. When the category score is calculated from the ALS algorithm, matrix factorization is performed using it. At this time, the maximum rating value is set to 1, resulting in a rating between 0 and 1. Additionally, because the factorization operation itself generates many recommended categories, only the top 100 are filtered by considering the rating value.

Afterwards, the control unit 120 calculates category score information through the ALS algorithm, combines the calculated category score information with the category information of the main product, and prioritizes the results.

An example of the category score being calculated for user u1 through the ALS algorithm and Matrix Factorization calculation and then combined with the category information of the mobile open market main product is shown in Table 5.

Category id Main Category
category id Middle classification category id Subcategory category id Rating One One 32 24 0.7 20 2 89 - 0.4 55 One 37 24 0 44 3 31 - 0 60 7 22 - 0

Referring to Table 5, the control unit 120 combines the category score information calculated through the ALS algorithm for user u1 and the category information of the mobile open market main product and sorts them in rank order, based on the sorted recommended category product information. This makes personalized main product recommendations possible.

Figure 5 is a diagram showing a flow chart between the user terminal 200 and the product recommendation server 100 according to an embodiment of the present invention.

Referring to FIG. 5, when a non-logged-in user accesses the mobile open market through the user terminal 200 and performs one or more of product inquiry, wishlisting, addition to shopping cart, and purchase without logging in, the non-logged-in user The latest usage information is transmitted to the product recommendation server 100, and recommended category product information is generated through the series of processes described above. The product recommendation server 100 delivers the generated recommended category product information to the user terminal 200 and exposes it as the main product on the mobile open market screen, so that non-logged-in users can view the recommended category product information.

Figure 6 is a diagram showing the connection structure between the user terminal 200 and the product recommendation server 100.

Referring to FIG. 6, the user terminal 200 is connected to the product recommendation server 100, which is a server, through a network as a client. The most important component of the client/server model is the network. The background to the development of the client/server model was the rapid development of local area network (LAN) and wide area network (WAN) technologies due to the development of connection devices such as bridges, routers, and gateways. . Attempts have been made to standardize various protocols appropriate for the developed local area network and wide area network network communication environment. The most representative one is the OSI 7-layer model developed by the International Organization for Standardization (ISO).

OSI 7-layer is a standard model for protocols announced by the International Organization for Standardization, dividing the network into 7 independent functional layers: physical layer, data link layer, network layer, transport layer, session layer, presentation layer, and application layer. Even though these are different, they are designed to enable mutual communication if the protocol is designed for each layer. The interface between adjacent layers can be defined as a service from the lower layer to the upper layer. First, the physical layer is the lowest layer and is responsible for transmitting data received from the data link layer, the upper layer, in bits through a communication line. The physical media comprised include devices that generate electrical signals such as hubs and repeaters, and mechanical connection devices such as telephone lines, UTP, and coaxial cables to manage electrical, mechanical, and physical conditions to transmit bit strings. It is a layer that supplements. The data link layer is the second layer and connects the bits of the lower layer, the physical layer, in a logical group called a frame. This layer controls the flow of data by detecting and recovering errors that occur in the transmission path between systems during communication, and ensures the reliable transmission of a set of data. It is responsible for transmitting data frames from the upper layer, the network layer, to the lower layer, the physical layer, and, like all other layers, adds its own identification information to the front of the transmitted data. Devices in the data link layer include intelligent hubs, bridges, network cards, and adapters. The network layer is the third layer and manages the relay function to establish a communication line with the final end system at the end of the communication partner using various communication networks such as data networks and telephone lines, thereby ensuring data transmission between the final end systems. The transport layer is the fourth layer and provides reliable data transmission, and the main data transmission protocols use TCP and UDP. If the probability of error occurring in the communication network connecting the system is high, the error can be detected and the reliability of data transmission can be improved by following the error recovery procedure. If some data is not delivered correctly from the receiving device, it can be retransmitted or the error can be reported to the upper layer so that the higher layer can take necessary action. The session layer is the fifth layer and manages a logical structure for communication between two communicating applications. In other words, it provides the ability to create a session between two applications to communicate, set breakpoints in between, and resume communication from the most recently set breakpoint if a problem occurs in the session. The expression layer is the sixth layer and converts or encrypts the various expression forms of the application layer, the upper layer, into a general-purpose transmission method and compresses the data. Lastly, the application layer is the seventh layer and defines the services required between applications that communicate through a communication network.

When exchanging information by connecting different networks, a router is a device that reads the address of the recipient contained in the transmission information and transmits it to another communication network using the most appropriate communication channel. For example, it is mainly used when relaying different local area networks or connecting a local area network to a wide area network. It is an essential piece of equipment when accessing the Internet, and is a key communication device that serves to set up a route to transmit information in communication networks that operate with different protocols. In addition to the bridge function that simply connects communication networks, it recognizes other communication networks according to the route assignment table and assigns a route, and determines the destination in other communication networks or the communication network to which it is connected based on received packets to select the most efficient route among several routes. Select and send the packet. It performs various communication network management functions, such as controlling communication flow and configuring several auxiliary communication networks within the communication network. The advantage is that the communication environment can be set up, and the overall network performance can be improved by determining the routing method according to the management policy. In addition, since the communication method is automatically determined according to standard logic, maintenance is easy, and since it is not limited by the communication method, a large-scale communication network can be easily configured, and communication volume can be distributed along various paths.

A gateway refers to a device that enables communication between networks using different communication networks and protocols in a computer network. In other words, it is a network point that acts as an entrance into another network, and in a broad sense, it is a device that acts as a passageway between different types of networks.

UDP is a protocol included in the transport layer. When information is exchanged on the Internet, it is a communication protocol in which the sender unilaterally transmits data without going through a connection procedure with the receiver and without sending a signal to send data. It is an unreliable and non-connection-oriented protocol because the sending side cannot and does not need to check whether the receiving side has received data, and there is no error correction or retransmission function. Although it is less stable than TCP, it has the advantage of being much faster.

Meanwhile, the protocol mainly used in the client/server model is the TCP/IP protocol. The TCP/IP protocol was developed in parallel with the design of the OSI model, and is simpler and more flexible, so the TCP/IP protocol is mostly used in actual fields. The TCP/IP protocol consists of four layers: physical layer, Internet layer, transport layer, and application layer. The physical layer plays the same role as the physical layer of OSI layer 7, and the Internet layer plays the role of the data link layer and network layer of OSI layer 7. The transport layer functions as the transport layer and session layer of OSI layer 7, and the application layer functions as the presentation layer and application layer of OSI layer 7.

TCP, like UDP, is a protocol included in the transport layer. When exchanging information on the Internet, a connection is established before transmitting data, and the sender divides the data into several packets and transmits them by numbering them. The receiving side reassembles the received packets according to their numbers. If packet loss, duplication, or error occurs, it detects this and requests retransmission. TCP is a reliable and connection-oriented protocol because it delivers completed data to the user on the receiving end. Although it is slower than UDP, reliable data transmission is possible.

FIG. 7 is a diagram showing the result of the recommended category main product information generated by the control unit 120 being transmitted to the user terminal 200 of a non-logged-in user via the transmitter and receiver 110. If a non-logged-in user accesses the mobile open market at least once, the user identifier mid is stored in the cookie, and information is collected at the time of use when the non-logged-in user performs one or more of the following: product inquiry, wishlist, add to shopping cart, and purchase. It is stored in the storage unit 130. Later, when a non-logged-in user accesses the mobile open market, the user is identified using the user identifier mid stored in the cookie, and the recommended category main product for each user transmitted through the transmitter/receiver unit 110 is displayed at the top of the first screen. I recommend it. For example, if a non-logged-in user has recently viewed, added to a wish list, added to a shopping cart, or purchased a cotton pants product in the mobile open market, it will be displayed as a recommended category main product on the first screen when accessing the mobile open market in the future. Cotton pants or jeans products are exposed as the main product.

A person skilled in the art to which the present invention pertains will be able to recognize that the present invention can be implemented in other specific embodiments without changing its technical idea or essential features. Therefore, the embodiments described above are merely illustrative and should be understood as not limiting in scope. In addition, the flow charts shown in the drawings are merely exemplary sequential sequences to achieve the most desirable results in carrying out the present invention, and of course, other additional steps may be provided or some steps may be deleted. .

The technical features described in this specification and implementations that implement them are implemented as digital electronic circuits, computer software, firmware, or hardware including the structures described in this specification and their structural equivalents, or a combination of one or more of these. It can be implemented as: Additionally, an implementation that implements the technical features described in this specification may be a computer program product, that is, as a module of computer program instructions encoded on a tangible program storage medium for controlling the operation of or execution by a processing system. It may be implemented.

A computer-readable recording medium may be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composition of matter that affects a machine-readable radio wave signal, or a combination of one or more of these.

Meanwhile, in this specification, “device” includes all instruments, devices, and machines for processing data, including, for example, processors, computers, or multiple processors or computers.

A computer program, known as a program, software, software application, script, or code, may be written in any form of a programming language, including a compiled or interpreted language, or an a priori or procedural language, and may be a standalone program or module, component, or subprogram. It may be implemented in any form, including routines or other units suitable for use in a computer environment. On the other hand, a computer program does not necessarily correspond to a file in the file system, either within a single file provided to the requested program, or in multiple interacting files (e.g., storing one or more modules, subprograms, or portions of code). It may be stored within a file), or within a portion of a file that holds other programs or data (e.g., one or more scripts stored within a markup language document). The computer program may be implemented to run on multiple computers or one or more computers located at one site or distributed across multiple sites and interconnected by a wired/wireless communication network.

Meanwhile, computer-readable recording media suitable for storing computer program instructions and data include, for example, semiconductor memory devices such as EPROM, EEPROM, and flash memory devices, magnetic disks such as internal hard disks and external disks, and magneto-optical devices. It can include all forms of non-volatile memory, media, and memory devices, including disks and CD and DVD disks. The processor and memory may be supplemented by, or integrated into, special-purpose logic circuitry.

Implementations that implement the technical features described herein may include backend components, for example, data devices, or middleware components, for example, application devices, or, for example, users may use implementations of the subject matter described herein. It may be implemented in a front-end component, such as a web browser or a client computer with a graphical user interface capable of interacting with the computer, or in a computing system that includes any combination of one or more of such back-end, middleware, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication, for example, a telecommunications network.

Hereinafter, in addition to the content described above, a more specific example that can implement the components included in the main product recommendation method for a non-logged user described in this specification will be described in detail.

The non-logged user main product recommendation method described herein is partially implemented through means of executing computer software, program code, or instructions on a client device or a device related to a web-based storage system or one or more processors included in the device. Or it can be used as a whole. Here, the processor may be part of a computing platform such as a device, client, network infrastructure, mobile computing platform, stationary computing platform, etc., and may specifically be a type of computer or processing device capable of executing program instructions, code, etc. In addition, the processor may further include a memory for storing a non-logged-in user main product recommendation method, commands, codes, and programs. If the processor does not include memory, the non-logged-in user main product recommendation method, commands, and codes may be stored through a separate interface. It is also possible to access storage devices such as CD-ROM, DVD, memory, hard disk, flash drive, RAM, ROM, cache, etc. where the program is stored.

Additionally, the non-logged user main product recommendation method described herein may be used partially or entirely through a device, client, gateway, hub, router, or device running computer software on network hardware. Here, the software can be executed on various types of devices such as file devices, print devices, domain devices, Internet devices, intranet devices, host devices, distributed devices, etc., and the above-mentioned devices include memory, processor, computer-readable storage medium, It may further include an interface for accessing storage media, communication devices, ports, clients, and other devices through wired/wireless networks. Additionally, non-logged user main product recommendation methods, commands, codes, etc. may also be executed by the device, and other devices required to execute the non-logged user main product recommendation method may be implemented as part of a hierarchy associated with the device. .

In addition, the device can provide an interface to other devices, including without limitation clients, other devices, printers, database devices, print devices, file devices, communication devices, distribution devices, etc., and connections through the interface can be made through wired/wireless networks. It is possible to facilitate remote execution of programs. In addition, any of the devices connected to the device through the interface may further include at least one storage device capable of storing a non-logged-in user main product recommendation method, commands, code, etc., and the central processor of the device may be executed on a different device. Commands, codes, etc. can be provided to the device and stored in the storage device.

Meanwhile, the non-logged user main product recommendation method described in this specification can be used partially or entirely through a network infrastructure. Here, the network infrastructure may include devices such as computing devices, devices, routers, hubs, firewalls, clients, personal computers, communication devices, routing devices, etc., as well as separate modules capable of executing each function, as mentioned above. In addition to one device and module, it may further include storage media such as story flash memory, buffer, stack, RAM, ROM, etc. In addition, the non-login user main product recommendation method, commands, codes, etc. can also be executed and stored by any one of the devices, modules, and storage media included in the network infrastructure, and are used to execute the non-login user main product recommendation method. Other devices as needed may also be implemented as part of the network infrastructure.

Additionally, the non-logged user main product recommendation method described in this specification may be implemented with hardware or a combination of hardware and software suitable for a specific application. Here, hardware includes both general-purpose computer devices such as personal computers and mobile communication terminals and enterprise-specific computer devices, and computer devices include memory, microprocessors, microcontrollers, digital signal processors, application integrated circuits, programmable gate arrays, and programmable array organizations. It may be implemented as a device including, or a combination thereof.

The computer software, instructions, codes, etc. described above may be stored or accessed by a readable device, where the readable device is a computer component, such as RAM or ROM, that contains digital data used for computing during a certain time interval. Semiconductor storage, permanent storage such as optical disks, mass storage such as hard disks, tapes, drums, etc., optical storage such as CDs or DVDs, flash memory, floppy disks, magnetic tapes, paper tapes, independent RAM disks, large capacity removable from the computer. It may include storage and memory such as dynamic memory, static memory, variable storage, and network-attached storage such as cloud. Meanwhile, the commands and codes here include data-oriented languages such as SQL, dBase, etc., system languages such as C, Objective C, C++, assembly, etc., architectural languages such as Java, NET, etc., and application languages such as PHP, Ruby, Perl, Python, etc. It includes all the same languages, but is not limited thereto, and may include all languages widely known to those skilled in the art in the technical field to which the present invention pertains.

Additionally, the “computer-readable recording medium” described in this specification includes all media that contribute to providing instructions to the processor for program execution. Specifically, it includes, but is not limited to, non-volatile media such as data storage devices, optical disks, magnetic disks, etc., volatile media such as dynamic memory, and transmission media such as coaxial cables, copper wires, optical fibers, etc. that transmit data.

Meanwhile, configurations that implement the technical features of the present invention included in the block diagram and flow chart shown in the drawings attached to this specification represent logical boundaries between the configurations. However, according to an embodiment of the software or hardware, the depicted configurations and their functions are executed in the form of stand-alone software modules, monolithic software structures, codes, services, and combinations thereof, and can execute stored program code, instructions, etc. Since the functions can be implemented by being stored in a medium executable on a computer equipped with a processor, all of these embodiments should also be regarded as falling within the scope of the present invention.

Accordingly, although the attached drawings and their descriptions illustrate the technical features of the present invention, they should not be simply inferred unless a specific arrangement of software for implementing these technical features is clearly stated. In other words, various embodiments described above may exist, and since such embodiments may be partially modified while retaining the same technical features as the present invention, these should also be regarded as falling within the scope of the present invention.

In addition, in the case of a flowchart, operations are depicted in the drawing in a specific order, but this is shown to obtain the most desirable results, and such operations must be executed in the specific order or sequential order shown, or all illustrated operations must be executed. It should not be understood as something that must be done. In certain cases, multitasking and parallel processing may be advantageous. Additionally, the separation of the various system components of the embodiments described above should not be construed as requiring such separation in all embodiments, and the program components and systems described are generally integrated together into a single software product or integrated into multiple software products. It should be understood that it can be packaged.

As such, this specification is not intended to limit the invention to the specific terms presented. Accordingly, although the present invention has been described in detail with reference to the embodiments described above, those skilled in the art can make modifications, changes, and modifications to the embodiments without departing from the scope of the present invention. Transformations can be made. The scope of the present invention is indicated by the claims described later rather than the detailed description above, and the meaning and scope of the claims and all changes or modified forms derived from the equivalent concept are interpreted to be included in the scope of rights of the present invention. It has to be.

100: Product recommendation server
110: Transmitter and receiver
120: control unit
130: storage unit
200: user terminal

Claims (12)

a transceiver unit that receives usage information of the non-logged-in user from the non-logged-in user's user terminal and transmits the generated recommended category product information to the user terminal;
Based on the usage information, a category interest score that reflects the latest is calculated, a category score is calculated using the category interest score, and the recommended category product information is generated by combining the category score information with the category information of the mobile open market main product. a control unit that does;
a storage unit storing the usage information, the category interest score, the category score, category information related to the mobile open market main product, and the recommended category product information;
Product recommendation server including According to paragraph 1,
The product recommendation server, wherein the usage information is stored as information on the latest usage point of the non-logged-in user. According to paragraph 1,
The usage information is a product recommendation server characterized in that it includes mid (Member id) stored in a cookie when the non-logged-in user logs in at least once. According to paragraph 1,
The usage information is a product recommendation server characterized in that it includes one or more of the major, medium, and small categories in product classification. According to paragraph 1,
A product recommendation server wherein the number of product clicks is applied as a weight to the usage information of the non-logged-in user in the process of calculating the category interest score. According to paragraph 1,
The category score is a category interest score for each category calculated based on the collected usage information after collecting the usage information according to one or more of viewing, wishing, adding to the shopping cart, and purchasing the product by the non-logged-in user. A product recommendation server characterized by being calculated by inputting into the ALS (Alternating Least Square) algorithm. In the method of the product recommendation server recommending the main product,
(a) receiving usage information of the non-login user from the non-login user's user terminal;
(b) storing the usage information;
(c) calculating a category interest score reflecting recency based on the usage information;
(d) calculating a category score using the category interest score;
(e) generating recommended category product information by combining the category score information and category information of the mobile open market main product, and transmitting the recommended category product information to the user terminal;
Product recommendation method including delete delete delete delete delete

Images