KR100762829B1 - Intelligent searching service system and method in accordance with the customer?s personalized shopping preference - Google Patents

Abstract

The present invention relates to a product search system and method for presenting a plurality of search conditions to display a consumer's taste, preference, etc. in online shopping, and searching for products according to selected criteria to recommend products most suitable to consumers in real time. In the product group (for example, a set of specific products such as a mobile phone, a TV, wine, clothing, etc.), those that can define the attributes and characteristics of the product group are selected as search criteria and stored in the product group search criteria database. By defining the attributes (level or extent in the search criteria) of individual products in the product group, if the consumer presents his or her taste and preference for the product group, the consumer's preference is calculated by calculating the suitability between each product and the consumer's taste (preference). List the products that best match your taste in order of conformity. By doing so, the consumer can quickly find a product suitable for his or her taste, which could not be found by the conventional text-based search technique, and can reduce unnecessary shopping time.

Consumer, internet, merchandise, shopping, sale, recommendation, taste, preference

Description

Intelligent searching service system and method in accordance with the customer ’s personalized shopping preference}

1 is a block diagram of a product search system according to an embodiment of the present invention.

2 is an operation flowchart of a product search system according to an embodiment of the present invention.

3 is an example of an input window for registering an attribute of a product.

4A and 4B are examples of search boxes for inputting consumer tastes and preferences.

5A and 5B show an example of a search box of a real consumer.

6 is a screen showing a result of a consumer inputting and searching his or her taste in the case of women's clothing.

7 is a diagram illustrating an example of cumulatively reflecting an evaluation of a buyer after purchasing an actual product in the attributes of individual products.

The present invention relates to a product search service system and method for searching for and providing goods and services (uniform to the following) according to the customer's personalized taste in electronic commerce using the Internet.

E-commerce using the Internet has already been widely spread in real life and is being used by a large number of people already. It is expected that high growth will continue in the overall market size. There are numerous shopping malls and open markets (hereinafter referred to as shopping sites) on the Internet, and there is a fierce competition for online shopping customers. Accordingly, shopping sites are constantly making efforts to provide a better shopping environment and customer service. In particular, as the provision of personalized services according to the characteristics and tastes of customers is very important in internet shopping, various attempts have been made in various directions. Techniques such as customer relationship management (CRM) or 1: 1 marketing have been applied to provide differentiated services to customers, but so far, products that reflect personal preferences and product selection criteria are aimed at consumers. The recommendation service is still unsatisfactory. In addition, due to the nature of the Internet, which allows unlimited space use, each shopping site displays products ranging from thousands to tens of thousands of items on a flat web browser in a uniform manner such as sales order or popularity. As products are presented differently from their own tastes, they are experiencing information overflow. In particular, if you do not know the specific model name or brand, you can not search the text, but also if you want to find a product that reflects your personal sentiment, taste, and preferences for the purpose of the purchased product.

The technical problem to be achieved by the present invention is to solve the disadvantages of the prior art, to allow consumers to search for products reflecting personal preferences when shopping online, and to suit the relationship between the attributes of the product and the taste of the customer To provide a product search system and method that calculates (hit ratio) and displays search results.

Product search system according to one feature of the present invention for solving the technical problem,

A web page database for storing web pages for a search service;

A product database for storing product information;

A search criteria database for each product group that stores search criterion information for each product group that is the same kind of product;

A product-specific attribute information database for storing attribute information of the corresponding product determined according to the search criteria;

When a user terminal is connected, the webpage of the web page database is provided, and a search service unit receives a desired product group and attributes from the user terminal and provides corresponding products with reference to the product-specific attribute database and product database.

The search service unit,

A user interface unit which provides a web page of the web page database and provides a search result screen so that a user can search by product group together with a basic menu;

A product group search criteria management unit for registering a product group search criterion in the product group search criteria database according to an external input;

A product-specific attribute manager that registers attributes of individual products belonging to a specific product group in the product-specific attribute database according to external input;

A hit ratio calculation module that receives a product to be searched from a user terminal and a product-specific attribute according to a user's preference, compares the input attribute with an attribute of an individual product, and provides product information in order of decreasing difference;

And a consumer taste calculator configured to calculate an average value of attributes of goods selected by a user by a user terminal, and calculate calculated average attribute information so that the hit ratio calculation module calculates the hit ratio.

Product search method according to one feature of the present invention for solving the technical problem,

A product search method comprising a server and a database, and selecting a desired product from a user terminal to search and providing the result,

Receiving, by the server, a product group to be searched from the user terminal and attributes predefined in the corresponding product;

And comparing the input attribute information with attributes of the corresponding products belonging to the product group stored in the database, and providing a product having a difference in a certain range.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention. Like parts are designated by like reference numerals throughout the specification. In the exemplary embodiment of the present invention, a simple search method that does not include an individual's emotion or taste, such as a word-based text search, that is, a model name, a brand name search, or a price range search, is already used in general.

In the embodiment of the present invention, a product group is defined as a concept of a set of similar products such as TV, digital camera, MP3 player, audio, mobile phone, clothing, handbag, shoes, furniture, travel, movie / performance ticket, and the like.

In addition, the search criteria for each product group means that the functions or characteristics of a specific product group are defined so that ordinary consumers, not experts, can easily judge. For example, in the case of a mobile phone, a photographing function, a music listening function, a video recording function, a text transmission function, a screen quality level, a voice call function, and other additional functions may be search criteria of a product group called a mobile phone. In another example, in the case of wine, criteria such as sweetness, bitterness, no aroma or richness may be search criteria for a wine product group. Some consumers prefer sweet, fragrant wines, while others may prefer slightly bitter, slightly flavored wines. This difference plays an important role in the present invention as a personalized product search criteria used when recommending products that meet the consumer's product selection criteria. Next, product-specific attributes define the level or degree of individual products in the search criteria of the product group. For example, certain wines are very sweet and richly flavored.

1 is a block diagram of a product search system according to an embodiment of the present invention.

Referring to FIG. 1, a product search system according to an exemplary embodiment of the present invention includes a search service unit 300 and a database unit 400.

The product search service unit 300 includes a user interface unit 310, a product group search criteria management unit 320, a hit ratio calculation module 330, a product attribute management unit 340, and a consumer taste calculator 350. The user interface 310 provides a search box for a personalized search for each product group along with a basic menu to a user terminal 100 connected to a shopping site server, and provides a search result screen according to a consumer's taste. . The search criteria management unit 320 for each product group plays a role of registering and updating and managing search criteria that can express consumer's taste when a general consumer purchases a corresponding product for each product group such as TV, digital camera, MP3 player, and computer. do. The product attribute management unit 340 registers, updates or manages the attributes of individual products belonging to a specific product group. The attributes of individual products may be expressed differently according to search criteria of a specific product group. In addition, the hit ratio calculation module 330 is a module that calculates the hit ratio through a statistical suitability analysis by comparing the attributes of the individual products belonging to the product group and the customer's taste for the product group to recommend the customer's taste. The consumer taste calculating unit 350 calculates the consumer taste by calculating an average of attribute values of a predetermined number of products of the product group selected by the user.

The database unit 400 includes a member database 450, a web page database 440, a search criteria DB 410 for each product group, a product DB 420, and an attribute DB 430 for each product. The member database 450 stores a member's ID, password, personal information, product information selected by the member, and attribute information for each product group. Webpage database 440 stores webpages for wives and services. The search criteria database 410 for each product group stores search criteria for defining a consumer's taste for each product group such as a TV, a digital camera, a mobile phone, and clothing. For example, in the case of a mobile phone, the search criteria store the basic uses of the mobile phone as search criteria such as a photographing function, a music listening function, a video function, a text transmission function, and a voice call function. The search criteria for fashion products such as clothing, shoes, and handbags can be classified emotionally, such as romantic style, modern style, manic style, and natural style. The merchandise DB 420 stores all data about the merchandise sold at the shopping site. The product-specific attribute DB 430 stores the attributes according to the search criteria of the individual products stored in the product DB 420. For example, some wines are very sweet and slightly fragrant, with attributes stored for that wine. In this case, the property that is very sweet or slightly dark is actually stored in the DB as a numerical value applying the concept of 0 to 10 or a few percent (%).

The operation of the product search system having the above configuration will be described in detail with reference to FIGS. 1 to 6.

2 is an operation flowchart of a product search system according to an embodiment of the present invention.

Referring to FIG. 2, first, an expert or a consumer group registers a search criterion using a terminal or other device for a function or use that can best define a specific product group (S100). In this step, the search criteria management unit 320 inputs the search criteria such as the function, use, and feature classification of the product group (for example, a romantic style in the case of fashion products) by the operator computer or other means for each product group. Receive and register. Consumers play a very important role when buying products from a particular product family. As mentioned earlier, in the case of a mobile phone, a photographing function, a music listening function, a video recording function, a text transmission function, a screen quality level, a voice call function, and other additional functions are important factors in the purchase, and the search criteria are clothes or handbags. In the case of a fashion product such as the fashion style of the product (cute and lovely romantic style, stylish modern modern style, comfortable and generous natural style, etc.) may be the search criteria of the fashion product group. In the case of electronic or IT products, the functional development is so fast that new functions must be added and changed, and in the case of fashion, the addition or change of the newly expressed fashion style must be made continuously. Can be. Here, various methods may be applied to select search criteria of a specific product group. Extract search criteria to define the characteristics, functions, and uses of a specific product group from a group of experts in the field, or form a general consumer group to survey or interview factors that are most important when purchasing a specific product group. You can also converge the search criteria of a specific product group through. In addition, there may be various methods, which may be applied differently according to the product group by dividing the case where the expertise is necessary and the case where it is not. The search criteria for each product group may be changed to reflect the development of technology or a change in taste, or may have different weights according to the search criteria through analysis of data mining.

If the search criteria for each product group is registered, next, the level or degree for each search criteria of the individual products belonging to the corresponding product group is registered (S101). This is because the attributes of individual products can be searched according to this search criterion. Therefore, the attribute data of each product, that is, the level or degree of the corresponding product for each item of a specific search criterion is previously input before the search and stored in the database unit 400. The registration of an attribute of an individual product may be different according to a search criterion. In case of a somewhat objective search criterion, for example, a certain criterion is input according to the number of pixels of a camera in the case of photographing function in a mobile phone. In addition, when the judgment of several people is more important than the judgment of any one person, such as emotion or individual judgment, consumers who actually purchased can evaluate the attribute and input more converged information. In the registration of the attributes of individual products belonging to a specific product group, the seller initially registers the attributes of the product according to the criteria presented when registering the search criteria of the product group. When registering or evaluating an attribute of a product according to a search criteria of a product group, a user should provide an input window in the same form as a search box in which a consumer presents his / her taste as shown below. Only then can comparisons be made under the same criteria in the conformity review.

3 shows an example of a window for inputting an attribute of a product.

Referring to FIG. 3, an operator or a product seller may input a preference for each item of a search criterion in a unidirectional or bidirectional manner using an input window as shown in FIG. 3 displayed on the screen of the operating terminal 100.

When the preparation for the product search according to the consumer's preference is completed as described above, the consumer accesses the product search system using his terminal 200 to find a desired product (S102).

Then, the user interface unit 310 provides a web page of the web page DB 440 without going through a member authentication procedure or the need for such a procedure by referring to the member database 450.

Next, the user inputs his or her taste (preference) through the search input window according to the search criteria of the product group he wants to find using the terminal 200 (S103). In this case, the search input window is provided in a manner other than a text window for inputting words to express tastes or preferences. 4 shows an example of a screen for inputting a preference or a preference for a function under a corresponding search criterion when the search criteria of a specific product group are determined.

To express one's own needs or preferences for style or taste, existing text-based input is impossible. Therefore, a new expression method is required for inputting search information. Referring to FIG. 4A or 4B, in the present invention, a search bar such as a slide bar or a section scale method can freely express its emotion or preference. To provide. The reason for using such a search box is that consumers' emotions or preferences for functions cannot be accurately defined in the form of text. As shown in FIG. 4A or FIG. 4B, the search box may be implemented as a unidirectional expression or a bidirectional expression as an example of a slide bar type asking a customer's preference and a selection method according to a section scale. As a specific example reflecting the taste of the customer, FIG. 5A shows an example of a cell phone and a wine. First, a search box of consumer A and consumer B for a mobile phone product is illustrated. In search example 1, consumer A prefers a product having excellent music listening and voice call function among mobile phone functions, and consumer B of search example 2 takes a picture. It should be noted that the features and ringtones should be very good, and that you are looking for a product that has excellent texting capabilities. Next, search examples 3 and 4 of FIG. 5B show the tastes of consumers C and D when the product group is wine. It can be seen that Consumer C prefers wines that are very sour and fragrant, and Consumer D prefers wines that are very sweet and slightly flavored. As described above, consumers can express their tastes based on preference or importance rather than text, and these preferences are referred to as attributes.

As described above, the user's taste may be input through the search box. If necessary, the consumer taste calculator 350 may click the user's click flow when the user clicks a plurality of products on the shopping site. flow) by analyzing the average of each item's preference of items clicked to calculate the average value of the attribute, and can also provide an automated search function based on click analysis to search the average attribute by the search criteria . For example, if a consumer clicks on A, B, or C products while shopping in the MP3 player family, the system automatically averages the attributes of the product (the average of the attributes of A, B, and C products) to view similar products. Provides a button to calculate the average value of the items clicked by the consumer as the consumer's taste, and compares the property value with the property value of the product group, and the difference is within a certain value or a certain number of products. 330 may search and provide. When analyzing the click flow of the customer and showing the products that the consumer wants to find, the consumer has the benefit of automatically searching for the desired product even if there is no separate input through the search box.

When the consumer's taste is input from the terminal 200, the hit rate calculation module 330 compares and analyzes the attributes of the products belonging to the product group with the consumer's taste (S104). Here, the hit rate calculation module 330 compares the suitability between the attributes of the goods already registered and the taste of the consumer, and calculates the hit rate with the consumer taste for each product. In order to calculate the customer's taste, that is, the suitability between the product selection criteria and the attributes of individual products, for the same search criteria of a specific product group, the following formula for calculating the suitability may be applied. In addition, various formulas for calculating the suitability may be applied. Can be. Here, the illustrated formula is a method of calculating the absolute distance value by calculating the absolute distance of two different points in the N-dimensional space.

Assuming that there are n variables (V), which are search criteria for a specific product group, the tastes of customers are set to V11, V21,... , Vn1, and V12, V22,... , Vn2, the suitability between customer's taste and product attributes is the absolute distance between two points in n-dimensional space, where the small absolute distance value is high, and the large absolute distance value is low mutual compatibility. Can be. In addition, each search criterion may be weighted, and the more important the search criterion is, the higher the weight may be calculated.

The taste of the customer and the absolute distance value of the corresponding product weighted by the search criteria may be expressed by Equation 1 below.

In Equation 1 above, W1, W2,... Wn is a weight that can be assigned to each search criterion and can be adjusted according to the characteristics of the individual search criteria. In addition, when the consumer does not display his or her taste in the search box, the weight should be zero. This application will depend on the nature of the product group and which search criteria are used. In addition, according to the above absolute distance value, the hit ratio (%) may be calculated by Equation 2 below.

Customer's taste and hit rate of the product

= 100 * (Absolute distance value between customer's taste and the corresponding product / theoretically maximum absolute distance value) Here, theoretically, the maximum possible absolute distance value is calculated in advance and stored for each product.

The customer's taste and the attributes of the individual products are calculated and compared using the above formula, and the hit ratio calculation module 330 prefers the products having a small absolute distance value, that is, the products with a high hit rate that match the customer's taste. To provide. By showing the personalized product search results, customers can not only increase their satisfaction, but also have the advantage of finding their desired products in a short time. This process is expected to ultimately lead to an increase in the conversion rate of visitors to the shopping site.

Then, the terminal 200 receives and displays a product according to the hit ratio calculated for each product from the hit ratio calculation module 330 and displays it to the consumer (S105). In this step (S105), when the consumer enters his or her tastes in the search box according to the search criteria of a specific product group, the hit ratio calculation module 330 compares and analyzes the attributes of the individual products belonging to the product group to calculate the suitability, The closest product in suitability is selected first, and the search result screen is shown in order of suitability (hit ratio). An example of women's clothing for this is shown in FIG.

Referring to FIG. 6, when a consumer inputs a fashion style he or she is looking for in a women's clothing product through a search box, the hit rate calculation module 330 compares all the attributes of each women's clothing product and the input consumer taste. The products with the highest hit rate that best meet the consumer's taste are shown in order.

If you show all products, you can expose all products from 100% to 0%. When you show the search results, if you apply the filtering option in the search to exclude products with a hit rate of less than 50% Only a limited number of more suitable products can be presented. These filtering conditions may be registered in advance in the system as needed, and may allow the consumer to determine the number of recommendations in advance.

Next, the consumer may search based on the search result or search for the product by searching with another search criteria (S106). When the consumer selects a purchase, the user interface 310 processes a purchase request, and this purchase process is well known in this field, and thus a detailed description thereof will be omitted.

In this way, consumers can easily search for products that match their preferences and expose products in order according to the hit rate, thereby providing a fast and personalized service to consumers. On the other hand, the seller is expected to be able to increase the conversion rate of the product by exposing the product that suits their products and consumers' taste.

Thereafter, when the online purchasing process is completed, there may be an evaluation of the person who finally purchased. That is, when several people purchase and re-enter the attribute of the purchased product for each search criterion using the terminal 200, the attribute management unit 340 for each product changes the attribute of the corresponding product with reference to such input information. It may also be (S107).

In particular, if the emotional feelings are important products or the evaluation of multiple customers can have a significant impact on other people's purchasing decision, the consumers who purchased the actual products can be evaluated so that the attributes of the products can be continuously changed. Should be. At this time, when the product attribute is changed, the relative weight of the seller and the consumer may be managed by giving a relative weight. If the attributes of the product are re-evaluated, the attributes of each product are changed.

7 shows that after the actual purchase of a fashion product by the consumer, the attributes of the product change according to the product evaluation. Referring to FIG. 7, product attributes may be changed by reflecting evaluation values of evaluators in a cumulative average.

In the present invention, since the attributes of the product may be continuously changed according to the evaluation of the buyer, it may reflect a change in sensitivity.

The embodiment of the present invention is only one embodiment, and various modifications and changes are possible without departing from the gist of the present invention, and the present invention is not limited to the embodiment.

As described above, in the embodiment of the present invention, it is possible to provide a real-time product search system and method according to the taste of the consumer having the following effects.

First, there is an advantage in that the consumer can find a product that suits his taste above all. Existing product search methods extract only results that are limited to model names or brand names based on texts, but could not provide any search results for the emotions or preferences of consumers. It is configured to search from various points of view such as the importance of the use of the product can provide a differentiated search service. In particular, it is expected to have an excellent effect in emotional fields such as fashion, food and beverage, and movie / performance tickets, which have strong emotions and tastes of consumers.

Second, the merchandise can be found quickly according to personalized search. In many existing shopping sites, consumers have to read all the descriptions of products while searching a myriad of products in a specific product group, but in the present invention, by providing limited information to consumers, only the products they want can be quickly searched and purchased. You can get time benefits.

Third, the shopping site has the advantage of improving the conversion rate by searching for and recommending the right product to the consumer. The purchase conversion rate may be defined as a purchase rate per visited customer, and as a recommendation of a more suitable product to a consumer is expected, the likelihood of purchase is expected to increase, and thus, the present invention is expected to improve the purchase conversion rate of a consumer.

Claims (10)

A product search method comprising a server and a database, and selecting a desired product from a user terminal to search and providing the result, Receiving, by the server, a product group to be searched from the user terminal and attributes predefined in the corresponding product; Comparing the received attribute information with attributes of the corresponding products belonging to the product group stored in the database, and providing a product having a difference in a certain range; The attribute information is a value selected in an analog form for each item for a predetermined number of search criteria indicating a corresponding product, Each item of the attribute is weighted according to importance, And using a hit ratio as a value for comparing the input attribute information with attributes of the corresponding products, wherein the hit ratio is calculated by the following equations (1) and (2).  [Equation 1]  Absolute distance between customer's taste for specific product group and attributes of individual product A

 [Equation 2]  Customer's taste and hit rate of the product = 100 * (Absolute distance value between customer's taste and the product / theoretically maximum absolute distance value) delete The method of claim 1, Registering search criteria such as emotion or function for a product group in a database in advance; The product search method further comprises the step of registering in advance the attributes of the individual products in accordance with the search criteria of the product group. The method according to claim 1 or 3, And evaluating the attribute after the user purchases the provided product, reflecting the result in the attribute of the corresponding product stored in the database. delete A web page database for storing web pages for a search service; A product database for storing product information; A search criteria database for each product group that stores search criterion information for each product group that is the same kind of product; A product-specific attribute information database for storing attribute information of the corresponding product determined according to the search criteria; When the user terminal is connected, provides a web page of the web page database, and receives a desired product group and attributes from the user terminal includes a search service unit for providing the corresponding product by referring to the product-specific attribute database and product database,  The attribute information is a value selected in an analog form for each item for a predetermined number of search criteria indicating a corresponding product, and is characterized by weighting each item of the attribute according to importance. The search service unit, A user interface unit which provides a web page of the web page database and provides a search result screen so that a user can search by product group with a basic menu; A hit rate calculation module that receives a product to be searched from a user terminal and a product-specific attribute according to a user's preference, compares the input attribute with an attribute of an individual product, and provides product information in an order in which the difference is small; The hit ratio calculation module calculates a hit ratio by the following equations (1) and (2).  [Equation 1]  Absolute distance between customer's taste for specific product group and attributes of individual product A

 [Equation 2]  Customer's taste and hit rate of the product = 100 * (Absolute distance value between customer's taste and the product / theoretically maximum absolute distance value) delete The method of claim 6, The search service unit, And a consumer taste calculator configured to calculate an average value of attributes of goods selected by a user by a user terminal, and calculate calculated average attribute information so that the hit ratio calculation module calculates the hit ratio. The method of claim 8, The search service unit, A product group search criteria management unit for registering a product group search criterion in the product group search criteria database according to an external input; According to an external input, the property of each product belonging to a specific product group is registered in the property database for each product, and after the purchase is complete, the product-specific attribute management unit characterized in that for changing the attribute of the selected product according to the selection of the user terminal Commodity Search System. delete

Images