KR102244563B1 - Contents matching method using rasch model-based intelligent db technics - Google Patents

Abstract

By measuring the item ability for each user and matching content with other users who have the appropriate ability based on the measured ability, content can be matched with users who have the optimal ability for the user, thereby increasing the matching satisfaction. In particular, by measuring the user's ability by applying the Rasch Model with the level or frequency input by each user as a variable, after measuring the abilities of each user through variables, users or contents with similar levels can be matched with each other, thereby further enhancing the matching satisfaction. In addition, by configuring items to be upgraded on a regular, daily, weekly, or monthly basis, the matching effect can be further enhanced even for items which can change frequently in a short time, such as fashion, design, and artistic performances.

Description

Content matching method using intelligent DB generation technology based on Rash model {CONTENTS MATCHING METHOD USING RASCH MODEL-BASED INTELLIGENT DB TECHNICS}

The present invention relates to a content matching method using an intelligent DB generation technology based on a Rasch model, and more particularly, by applying a Rasch model to an item that can measure capability among items entered by a user. By measuring the user's ability for each item, it becomes an intelligent DB, and when the user selects an item to match, the user can select from users and contents with the ability of the matching range determined in advance based on the user ability of the item. It is possible to further enhance the matching effect by matching with other users or content with the most appropriate ability for the company's ability. At this time, this intelligent DB is configured to be upgraded on a regular, daily, weekly, or monthly basis, thereby further enhancing the matching effect.

In general, as travel is made up of individual trips, many matching methods have been developed to find locals to travel with in the region or to experience various cultures while traveling with travelers of different nationalities. In the following (Patent Document 1) and (Patent Document 2), techniques relating to a method of making friends using a network are disclosed.

(Patent Document 1) Korean Patent Registration No. 10-1355639

The present invention relates to a method for introducing a friend and making a friend, and a system for the same, which enables the formation of a new network with other users within a limited range by using a relationship-based network. For the first user, each of the first user and a second user who is a friend of the first user receives status information that alternatively represents one of a plurality of different levels of status, and based on the received status information, A third user who determines the state of the first user to the second user as either public or private, is a friend of the second user, and has a serial connection relationship with the first user with the second user as an intermediary, and For at least one of the fourth user who is a friend of the third user and has a serial connection relationship with the first user with the second user and the third user as an intermediary, according to the state determined for the second user of the first user. Register the first user in the introduction pool. According to this, since the range of users that can form a network is determined by reflecting both the intentions of the users who want to make friends and the intentions of the intermediary, there is an advantage in that a reliable meeting is possible.

(Patent Document 2) Korean Patent Registration No. 10-0631708

Push-to-talk, a terminal that provides a push-to-talk service that allows you to introduce an anonymous friend after searching for a partner that meets the requested condition by using the subscriber information of the presence server that supports the push-to-talk service. It relates to a friend introduction system and method using a two-talk service. In the conventional PTT service, a call is possible only after obtaining the consent of a PTT user that he knows and registering it in the management server in advance, so there is no way to obtain information about anonymous users registered as PTT users. Calls are not possible, and club activities among PTT service subscribers or call attempts between anonymous users are fundamentally blocked, so there are many limitations in the use of PTT service. In consideration of this problem, by registering subscriber information in the management server that manages PTT subscriber information and setting whether or not to disclose the information, the management server can collect and search the disclosed user information, and the standard required by the user. By extracting matching targets from the disclosed user information, this is provided to a user who wants to introduce an anonymous user.

In addition, various matching technologies are applied and used worldwide as follows, but there are the following problems.

(1) Meetup: As the world's largest'activity' club matching service, it mainly helps to meet foreigners with the same interest as a group, but in Korea, only'language exchange' is active. It seems to have the same theme as the service, but it has not been able to solve the structural limitations of'group' matching and the risk of'no show'.

(2) Tinder: The service that has the most users in the world in the foreign friend matching sector, the intelligence information DB that the developer intends to proceed, and the data collection, analysis, and recommendation through user profile information in the recommendation algorithm sector, etc. Although it has the highest level of technology, it is uncomfortable to a large number of customers with needs for wellness tourism content because the theme is limited to'reverse dating', and there is no recommended technology according to the cumulative amount of activity of the user.

(3) Airbnb Trip: This is a tourism content sales and purchase service launched in 2016 by an'Airbnb' service operator that analyzes the relationship between 5 billion data related to accommodation conditions and reservations, calculates and recommends it using an algorithm based on modeling. There is a limit to the format of meeting locals and travelers, or in the form of purchasing services in which the user relationship is set as'seller' and'buyer'.

(4) Couchsurfing: A platform that has secured more than 15 million'travelers' customers around the world. The'Hangout' function allows travelers to find local friends, but it is technically inadequate, so it is limited to the level of the'bulletin board', and matching through a simple'comment' format along with the inherent limitation that the originating service was'stay sharing' It has technical limitations in that it is.

(5) Klook: A service that sells a variety of tickets, such as tourism content, activities, F&B, etc., all over Asia starting in Hong Kong. It is characterized by price competitiveness and high content quality, but it does not have a user community or information exchange function. It is recognized as a'write-and-delete app'.

(6) My Real Trip: A domestic (Korea) version of Klook, which sells local tourism content to Korean tourists traveling abroad, but can only be used when Korean tourists travel abroad, and is a content for inbound tourists. There are difficulties in activating tourism.

(7) WAUG: It is a tourism content reservation and purchase platform that focuses on Southeast Asia and has direct contracts with companies, but it does not strategically approach the aviation and lodging markets, and it is understood that it is only focusing on activities.

Korean Patent Registration No. 10-1355639 (Registration date: 2014.01.20) Korean Patent Registration No. 10-0631708 (Registration Date: 2006.09.27)

The present invention takes these points into consideration, and by measuring the ability of each user for an item requiring matching, and configuring the content to match other users with similar or superior ability based on the measured ability, Its purpose is to provide a content matching method using an intelligent DB generation technology based on the Rash model, which enables users with optimal capabilities to match content and improves matching satisfaction.

In particular, the present invention measures the user's ability by applying the Rasch Model using the level or frequency input by each user for the item as a variable in such a way to match items, thereby measuring at least one variable (for example, First, it is possible to measure the ability of each user through language level, interests, and disposition), and then match users or contents of similar level to each other. There is another purpose to provide a content matching method using intelligent DB generation technology.

In addition, the present invention configures the items used to measure the ability to be upgraded at all times, daily, weekly, or monthly, so that items that can change frequently in a short time, such as fashion and design, and art performances, are quickly Another object is to provide a content matching method using intelligent DB generation technology based on the Rash model, which is reflected in such a way that the matching effect can be further increased.

The content matching method using the intelligent DB generation technology based on the Rash model according to the present invention for achieving this purpose includes: a first step of classifying items requiring matching among data input from a user for a certain period of time; A second step of applying an item classified in the first step to a Rasch model for each item to measure capabilities for each user to form an intelligent DB; A third step of allowing the user to select at least one of the classified items, extracting other users or content in the matching range based on the user's ability from the selected item; And a fourth step of allowing the user to select at least one user or one content from among other users and contents extracted in the third step; The Lash model used in the second step is characterized by using the following equation.

In particular, the item that needs to be matched is characterized in that at least one of networking, lifestyle, culture and art, language used, disposition, food and beverage, activity, industry, travel, movie, music, and exercise and sports.

In addition, the matching range is set to a value added or subtracted by a predetermined range based on the user's ability, or set to be higher or lower than the user's ability.

Finally, in the first step, the item is configured to be updated on a regular, daily, weekly, or monthly basis.

The content matching method using the intelligent DB generation technology based on the Rash model according to the present invention has the following effects.

(1) Among the items entered by the user, items that can measure individual ability for all subscribers are quantified, and other users or content with items that match the user's ability to match the items selected by the user By matching, it is possible to further increase the matching effect.

(2) In particular, the matching item does not simply match the same user or content, but for example, the level of the matched item, and travel, for example, is not simply a backpacking trip that has been subdivided within the main category, or a package tour. It is possible to further enhance the matching effect by checking whether or not you like the city tour and matching users with similar capabilities or within a certain range.

(3) On the other hand, by using the Rasch Model, the ability of these matching items can be measured with only one variable, so it is possible to easily measure the user's ability for each matching item and convert it into an intelligent DB. It is convenient.

(4) In addition, by configuring such matching items to be updated on a regular, daily, weekly, or monthly basis, users' abilities are quickly updated according to rapidly changing trends such as lifestyle, art performances, and design. Thus, it is possible to find a matching partner or content that best suits the user's ability.

1 is a flowchart illustrating a content matching method using an intelligent DB generation technology based on a Lash model according to the present invention.
[Fig. 2] is a large classification table showing items that can be used for matching among items input by an input user according to the present invention.
[Fig. 3] is a table showing a result of detecting capability through a Rasch model for an item input by an input user according to the present invention.
[Fig. 4] is a table in which data extracted according to networking is selected as a matching item in the table of [Fig. 3].

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the accompanying drawings. Prior to this, terms or words used in the specification and claims should not be construed as being limited to their usual or dictionary meanings, and the inventor should appropriately define the concept of terms in order to describe his own invention in the best way. It should be interpreted as a meaning and concept consistent with the technical idea of the present invention according to the principle that it can be.

Therefore, the embodiments described in the present specification and the configurations shown in the drawings are only one of the most preferred embodiments of the present invention, and do not represent all the technical spirit of the present invention, and various equivalents that can replace them at the time of the present application And it should be understood that there may be variations.

[Composition of content matching method]

The content matching method using the intelligent DB generation technology based on the Rash model according to the present invention, as shown in [Fig. 1] to [Fig. 4], classifies items that need to be matched with other users among several items input by the user. Next, for each item, the Rasch model is used to measure the user's ability to create an intelligent DB, and match by configuring it to extract and select content with other users with appropriate capabilities according to the user's ability to match. It is possible to further enhance the matching effect by recommending other users or contents that match the capabilities of the item selected by the user who wants to do so.

Hereinafter, this configuration will be described in more detail with reference to the accompanying drawings. Here, the content matching method using the intelligent DB generation technology based on the Rash model according to the present invention is performed in four steps as follows, and thus, it will be described in detail by dividing it into steps.

end. Step 1

The first step is a step of classifying items requiring matching among several items input by a user for a certain period of time, as shown in [Fig. 1] and [Fig. 2]. That is, here, as an item that the user inputs, when explaining by taking the example of signing up for a site, the personal information, hobbies, and interests of the subscriber can be exemplified. Among them, personal information such as gender, age, and nationality that can be used for matching among subscribers' personal information, and hobbies and interests, are not only interested in each other, but also act as an important factor for matching. do. Accordingly, in the present invention, the items that can be matched among items input by different users are classified and converted into an intelligent DB, so that users can easily find other users or contents suitable for their items and capabilities.

Items requiring such matching can be used as long as at least two users show interest to each other and attract attention when searching for a matching target. In a preferred embodiment of the present invention, as such a matching item, as shown in [Fig. 2], as a major category, networking, lifestyle, culture and art, language used, disposition, food and beverage, activities, industry, travel, movies, music, and It is desirable to configure it to include at least one of exercise and sports. In addition, such matching items are a large category, and it is preferable to include a small category as follows. At this time, it will be readily apparent to anyone in the technical field to which the present invention belongs that the classification or naming of the large and sub-classes can be changed at any time according to the progress of the service.

1. Networking

2. Lifestyle

3. Culture and Arts

4. Language

5. Inclination

6. Food and beverage

7. Activity

8. Industry

9. Travel

10. Movies and music

11. Exercise and sports

As shown in Fig. 3, items that need matching are stored and managed as one intelligent DB, or each item is stored in each intelligent DB and integrated and managed, and each item is converted into an intelligent DB by measuring the user's ability. do.

In a preferred embodiment of the present invention, it is preferable to configure the items to be updated on a regular, daily, weekly, or monthly basis so that the ability evaluation for each item can be performed periodically. This is to further enhance the matching effect by configuring to be reflected in various items according to trends that change over time, such as music or design.

I. Step 2

The second step is a step of measuring and digitizing the user's ability for the items classified in the first step, as shown in [Fig. 1] and [Fig. 3]. At this time, the Rasch model is used to quantify the capabilities of the user's items. Here, the Lash model is a conventional technique for estimating the ability of a subject using only one variable. In the present invention, the ability is measured as shown in [Equation 1] below by using the number of users who have selected each item as a frequency variable. did.

In [Fig. 3], SD was obtained by applying the frequency as a variable to the Lash model for the sub-category item (kr_section_title) classified according to the large-category item (kr_category_title), and this value is used as it is as the capability of each sub-category item, or ( 1-SD) can also be used as an ability as a value.

All. Step 3

In the third step, as shown in [Fig. 1], if the user who wants to match selects the desired item from the result of measuring the individual's ability from the classified items, the selected item has the ability appropriate to the item ability of the user who is trying to match. This is the step of extracting other users or content.

For example, assuming that the user trying to match is trying to match other users who are interested in networking, first, if you select'networking', which is a large category (kr_category_title), several types of sub-category (kr_section_title) items are displayed as shown in [Fig. 4]. Is extracted. In this case, here, it is described that a large category and a small category are sequentially selected, but the present invention is not limited thereto, and a search function may be used to directly select and search a small category item from networking, which is a major category.

On the other hand, as shown in [Fig. 4], the sub-classified items extracted according to the selection of the large classification are not exposed to the user as they are, and the sub-classified items with appropriate capabilities are extracted and shown again in consideration of the user's capabilities. In other words, the matching range is determined based on the ability of the user to be matched, and users corresponding to the sub-classes within the range are shown as targets for matching. Here, the matching range is preferably set to a value added or subtracted by a predetermined range based on the user's ability, or set higher or lower than the user's ability. There may be items that are irrelevant even if there is a greater difference than the user's ability to match, such as travel recommendations, drama, and cuisine, and it is too different based on the user's ability to match, such as the language ability used. This is because there may be items that make it impossible to communicate or communicate between users. Accordingly, it is desirable to extract other users in an appropriate matching range based on the capabilities of the user for the item selected by the user to be matched.

In addition, in a preferred embodiment of the present invention, in the above-described matching method, the case when a user searches for another user is described as an example. As a result, it can also be used to find or search for other content that fits it.

la. Step 4

In the fourth step, as shown in [Fig. 1], the other user information or content extracted in the third step is provided to a user who tries to match, and the user who tries to match is randomly selected from among the various users and contents to be extracted. . At this time, the user who tries to match is selected from among various users and contents extracted according to his or her ability, considering his or her disposition or intention, so that satisfaction and matching effect can be further enhanced.

The present invention made as described above is shown to a user who is trying to match by extracting an appropriate target by comparing the capabilities of other users based on the ability of the user trying to match and converting it into an intelligent DB by numerically converting the field of interest of the individual. By giving, it is possible to obtain an optimal matching effect suitable for the field of interest of the user who is trying to match.

Claims (4)

A first step of classifying items requiring matching among data input from a user for a certain period of time;
A second step of applying an item classified in the first step to a Rasch model for each item to measure capabilities for each user to form an intelligent DB;
A third step of allowing the user to select at least one of the classified items, extracting other users or content in the matching range based on the user's ability from the selected item; And
A fourth step of allowing the user to select at least one user or one content from among other users and contents extracted in the third step; including,
The content matching method using an intelligent DB generation technology based on the Rash model, characterized in that the Rash model used in the second step uses the following equation.

In claim 1,
Items that need matching above,
Networking, lifestyle, culture and art, language of use, inclination, food and beverage, activities, industry, travel, movies, music, and sports and sports using intelligent DB generation technology based on the Rash model characterized How to match content.
In claim 1,
The matching range is,
Set as a value added or subtracted by a predetermined range based on the user's ability, or
Content matching method using intelligent DB generation technology based on Rash model, characterized in that it is set higher or lower than the user's ability.
In any one of claims 1 to 3,
In the first step, the content matching method using an intelligent DB generation technology based on a Lash model, characterized in that the item is configured to be updated on a regular, daily, weekly, or monthly basis.

Images