KR101503348B1 - Mobile advertising method and system using user information - Google Patents

Abstract

Provided are a mobile advertisement method and an advertisement system. In the method, user groups are classified by user type, a query inputted by a user is collected by user group to generate a group profile, a related keyword is extracted from the group profile, and advertisement content is scored from the group profile according to a matching value with the related keyword. Also, a user group is selected, and a keyword is extracted from the inputted query, and advertisement content scored in the selected user group is selectively exposed in regard to the related keyword of the extracted keyword. The system includes an advertisement storage storing advertisement content provided by an advertiser or advertisement provider; an advertisement server scoring and selecting the advertisement content, stored in the advertisement storage, from the group profile according to the matching value with the related keyword after the inputted query and a related keyword index extracted from the inputted query are connected to the group profile classified by user type; and a user terminal to which the advertisement content is exposed.

Description

TECHNICAL FIELD [0001] The present invention relates to a mobile advertising method and system using a user-

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mobile advertisement method and an advertisement system using a user's usage pattern and more specifically to a mobile advertisement method and an advertisement system using a user's usage behavior, .

Hosts (referred to as "Website hosts" or "ad consumers") of the websites on which the ads are provided are those that raise the maximum ad revenue without damaging their user experience . Some website hosts are raising advertising revenue through their interests. Advertisers pay for placing their ads high on the search results list, or they pay when users click on the ads.

Some online advertising systems use ad related information and document content related information to "match " ads to documents (and / or scoring for ads) containing content such as, for example, web pages. Such online advertising systems use the relevant information of both the candidate ads and the document to determine the score of each advertisement for the document.

Existing advertising systems expose advertising content associated with keywords searched by a user, or exposed ad content related to keywords associated with a user's previous purchased product or service. Also, the advertisement contents are matched and scored based on the information of the users who are provided with the advertisement, such that the advertisement contents are highly scored using the geographical information, and are exposed to the users.

SUMMARY OF THE INVENTION The present invention provides a mobile advertisement method and an advertisement system that can expose more targeted advertisement contents than simply displaying advertisement contents based on a personal usage behavior of a user exposed to the advertisement.

Another object of the present invention is to provide a mobile advertisement method and an advertisement system capable of analyzing a user's usage behavior and exposing a targeted advertisement to a user's type.

According to an aspect of the present invention, there is provided a mobile advertisement method and an advertisement system using a user's usage pattern.

The mobile advertisement method according to the present invention includes the steps of: classifying a user group according to a user type; collecting a query inputted by a user, classifying the user group into the group profile, and extracting a related keyword from the group profile The method comprising the steps of: scoring advertisement contents according to the associated keywords and matching values in the group profile; selecting a user group; extracting keywords from the inputted query; And selectively exposing the scored advertising content in the group.

In one embodiment of the present invention, the step of extracting the related keywords may include extracting keywords from the query, analyzing the keywords, and analyzing the keywords in the same query, the keywords in a continuous query of the same user, And expanding at least one of the classified keywords into a related keyword.

In another embodiment of the present invention, the step of scoring the advertisement content includes extracting a query associated with the associated keyword within the group profile, and comparing the advertisement content to an associated keyword based on the extracted query and the matching value. And < / RTI >

In another embodiment of the present invention, the user's query may include at least one of a search keyword, a product query, a purchasing behavior, a URL address of a site visited by a user, an email keyword, an application installation and an edit document keyword.

In another embodiment of the present invention, the step of selecting the user group may include inferring at least one of the user's gender, age range and region using the user's previous query.

In yet another embodiment of the present invention, the method may further comprise updating a user's query for the exposed advertisement content to a group profile of the selected user group.

The mobile advertisement system according to the present invention includes an advertisement store in which advertisement contents provided by an advertiser or an advertisement provider are stored, an associated keyword index extracted from the input query, a group profile in which the input query is classified by user type, And an advertisement server for scoring the advertisement contents stored in the advertisement store according to the association keyword and the matching value and exposing the selected advertisement contents to the user terminal.

In one embodiment of the present invention, the user terminal may include an advertisement publisher that exposes the selected advertisement in the advertisement server to a predetermined area of the display.

In another embodiment of the present invention, the advertisement server may extract a query related to the related keyword in the group profile, and may score the advertisement content on the associated keyword according to the extracted query and the matching value.

In another embodiment of the present invention, the related keyword index may be a set of related keywords extended to at least one of keywords in the input same query, keywords in the continuous query of the same user, and keywords classified in the related keyword.

In another embodiment of the present invention, the advertisement server may update the group profile by receiving a user query for advertisement content exposed to the user terminal.

As described above, according to the present invention, it is possible to classify a user group according to a user's usage behavior, to provide a keyword associated with a user using a group profile of a corresponding user group, and to provide an advertisement matched to a related keyword, The advertisement contents can be exposed.

In addition, a high advertisement effect can be expected by scoring the advertisement contents in a customized manner using keywords having high frequency of the same user group.

Furthermore, by updating the group profile of the user group by feeding back the user's query on the exposed advertisement content, the reliability of the score of the advertisement content can be enhanced.

1 is a diagram illustrating an advertisement system according to an embodiment of the present invention.
2 is a diagram for explaining a user and group profile according to an embodiment of the present invention.
3 is a diagram for explaining an advertisement method according to an embodiment of the present invention.
4 is a flowchart illustrating an advertisement content scoring method according to an embodiment of the present invention.
5 is a flowchart illustrating an advertisement method according to an embodiment of the present invention.
6 is a flowchart illustrating an advertisement method according to another embodiment of the present invention.
7 is a diagram illustrating an example of an actual advertisement according to an embodiment of the present invention.

The above-mentioned objects, features and advantages will become more apparent from the following detailed description in conjunction with the accompanying drawings. Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating an advertisement system according to an embodiment of the present invention.

1, an advertisement system according to an exemplary embodiment of the present invention includes an advertisement server 120 for selecting an advertisement and transmitting the selected advertisement on a network 100, an advertisement publisher 114 Includes a user terminal 110 that is stored in the form of a software development kit (SDK) or the like, and an advertiser 130 that provides advertisements by having an advertisement store 132 in which advertisement contents are stored. The advertisement server 120 includes an advertisement selection engine 122.

The user terminal 110 stores an advertisement publisher 114 capable of communicating with the advertisement server 120 in a software development kit (SDK) format or the like. When the user terminal installs an application or the like, the advertisement publisher 114 can be distributed at the same time in a form installed in the user terminal.

The advertisement server 120 includes an advertisement content index 126 for providing advertisement content information of the advertiser, a profile DB 128 in which a usage pattern of a user group is stored, a related keyword index 124 for providing related keywords highly related to the keyword, Lt; / RTI >

The user terminal 110 is classified into user groups according to types by the advertisement server 120 and usage patterns of the user groups are stored in the profile DB 128. The usage pattern may be analyzed in a user's query such as a user's search keyword, a query item, an installation application, a URL address of a site visited by a user, a purchase product or an application, or a keyword of an edit document. In the profile DB 128, a group profile is stored for each user group.

The related keyword may be extended from at least one of the same keyword in the input query, the keyword in the continuous query of the same user, and the keyword set to be associated with the same keyword in the query inputted as the keyword classified as the related keyword in the dictionary.

When the user's query is inputted from the user terminal 110, the advertisement server 120 determines the user group in the profile DB 128 using the user information inputted in the query.

The advertisement server 120 selects an associated keyword of the keyword included in the input query from the associated keyword index 124 and selects the advertisement content matching the associated keyword in the advertisement content index 126. The selection engine 122 of the advertisement server 120 may select the advertisement content that matches the associated keyword in the advertisement store 132. The matching value matching the associated keyword can be set by various algorithms.

The selected advertisement contents are exposed to the user terminal 110 through the network 100. The user terminal 110 displays the advertisement 116 on a predetermined area of the display 112 of the user terminal 110 by the advertisement publisher 114 (see FIG. 7). Unlike FIG. 7, may be provided in the form of a push advertisement. In addition, the advertisement 116 may be exposed to a web page, a game screen, or the like by an advertisement publisher (not shown) installed in a server of a service company without going through the advertisement publisher 114.

When a query such as a click action of the user for the advertisement 116 exposed to the user terminal 110 is input, the publisher 150 selects the information content 150a, b, c ... for the advertisement, And displays it on the user terminal 110.

The search engine 140 displays the search result on the user terminal 110 through the search index 142 in response to a query such as a user's input of a search keyword.

A user query, such as a click on the ad 116 and a search on the search engine 140, is used to update the profile DB 128 with user information.

2 is a diagram for explaining a user and group profile according to an embodiment of the present invention.

2, a user query 202 such as a search keyword, a product inquiry, an email, an application installation, a purchase operation, an editing document, a URL address of a site visited by a user, location information, As a profile for the user profile 200 database. The individual user profile may include a set of keywords in the query 202, and the set of keywords may be ranked according to frequency.

The individual user profiles 200 are stored in a group profile 204 according to the type of user group, and each group profile 204 is aggregated to form an overall database 206. [

A group profile 204 may be generated based on the user's gender, age range, region, and the like. The group profile 204 may be analyzed from queries of user groups classified according to the user type. For example, queries of female users are collected and statistical analysis such as retrieval keywords, retrieval items, installation applications, purchase goods or services, extracting keywords of e-mail or editing documents, The property value can be deduced.

3 is a diagram for explaining an advertisement method according to an embodiment of the present invention.

Referring to FIG. 3, the advertisement publisher of the user terminal collects a query inputted by the user (302). The query can be a user's search keyword, a product inquiry, a buying act, an email keyword, an installation application, a keyword of an edit document, a URL address of a site visited by a user,

The collected query is transmitted to the advertisement server by the advertisement publisher, and the advertisement server generates the group profile by classifying the collected queries according to the user group (304). User groups can be grouped by user preference. For example, user groups can be grouped by gender, age range, region, and the like. Users' queries of each user group can be collected and the usage pattern of the user group can be configured as a group profile.

A keyword is extracted from the query input by the user terminal and the associated keyword is expanded (306). The related keyword may be determined from keywords entered in advance in the related keyword, or keywords displayed simultaneously or continuously in the same query.

And scored ad content that matches the associated keyword. For example, the query related to the related keyword may be extracted from the group profile and the advertisement content may be scored according to the matching value with the extracted queries (308).

Through the above processes, the targeted advertisement contents based on the user's usage behavior can be scored and prepared.

The step of exposing the ad at the user terminal begins at query input 310 of the user.

If a query is entered at the user terminal, the keyword may be extracted from the input query and the scored advertisement for the expanded associated keyword may be selectively exposed (312). An advertisement having a high score can be estimated as a user who intends to purchase the advertisement, so that the advertisement can be placed at the top. Here, a plurality of exposed advertisement contents may be exposed, the exposed advertisement may be presented in a list format, or may be repeatedly rolled in the same space at regular intervals.

For example, if a 20-year-old woman in A area searches for 'fashion hat' in search engine, 'heel', 'handbag' expanded in the intersection keyword set of A region group profile, 20 group profile, Etc. are presented as related keywords, and advertisement contents matching 'high heel' and 'handbag' may be exposed. At this time, the matched advertisement contents may be advertisement contents scored according to queries and matching values related to the related search terms in the A region, the twenties, and the user group to which the woman belongs.

As another example, if a man in his 30s uses a game called 'Name', a game in the same category as 'Name' is expanded to an associated keyword in the intersection of the B region group profile, the 30th group profile, and the male group profile And the game matched with the associated keyword can be exposed as advertisement contents. The game matched with the related keyword may be a game scored in the game category based on the frequency extracted from the intersection query of the B region group profile, the 30th group profile, and the male group profile.

4 is a flowchart illustrating an advertisement content scoring method according to an embodiment of the present invention.

Referring to FIG. 4, a user group is determined for each user type by collecting sex, age, area information, and the like of a user who inputs a query (402). The user group can extract the age range and gender of the user from the user information inputted in advance, and the local information can extract the connected region of the user through the geographic information system or extract the local information included in the query.

(404) a user profile including a search keyword, an inquiry item, an installation application, a purchase item or service, a keyword of an edit document, and a URL address of a site visited by the user. The user profile reflects the user's usage behavior.

The user profiles are classified into the user types and stored as the group profiles (406). If the user group is classified based on sex, the group profile can be classified into a male profile and a female profile. When the user group is classified based on the age group, the group profile can be divided into 10, 20, 30, 40 ... profiles. If the group profile is classified based on the region, it can be classified into a group profile of a certain radius region of the country, the administrative district, or the current access region.

Indexed (408) associated keywords that are concurrent or related to the keywords of the user profile. In the case of a set of keywords contained in the same query used in the profile, it can be concluded that the set of keywords contained in a continuous query of the same user has concurrency and has a correlation. In addition, it is judged that the keywords inputted in advance as related keywords are highly correlated.

A group profile may be utilized to score targeted advertising content. That is, the related keywords of the keywords included in the query may be expanded for each user group, and the rank may be designated by scoring the advertisement contents according to the related keywords and the matching value.

For example, the user's query related to the associated keyword may be extracted from the group profile to determine an advertisement category having a high matching value with the extracted query, and the advertisement content may be scored in the category according to the extracted query and matching value (410). This process is advantageous in that the targeted advertisement contents can be scored by reflecting the interest goods or services and the purchase behavior in the query of the user group without determining the advertisement contents according to the related keywords and the matching value.

5 is a flowchart illustrating a mobile advertisement method according to an embodiment of the present invention.

Referring to FIG. 5, an advertisement server collects a query inputted by a user terminal through an advertisement publisher installed in a user terminal.

In addition, a user group for each gender, age group, and region is defined, a group profile is generated according to a user query in the group, and a related keyword is extracted in a user query in the group (502). The user's query includes search keyword, query item, installation application, purchase product or service, keyword of the edit document, URL address of the site visited by the user, and the like. It is possible to extract keywords from this user query and to extract the related keywords by extending the extracted keywords. The related keywords may be extended from at least one of keywords in the input same query, keywords in a continuous query of the same user terminal, and keywords input in association keywords in advance.

The query related to the related keyword is extracted from the group profile, and the advertisement content is scored (504) for the associated keyword of the user group according to the extracted query and the matching value. The user's query related to the related keyword may be extracted from the group profile to determine an advertisement category having a high matching value with the extracted query and the advertisement content may be scored in the advertisement category according to the extracted query and the matching value. This process is advantageous in that the targeted advertisement contents can be scored by reflecting the interest goods or services and the purchase behavior in the query of the user group without determining the advertisement contents according to the related keywords and the matching value.

A user group to which a user who is a target of advertisement exposure belongs is selected (506). In the present embodiment, the user group selects a user group including at least one of the user's sex, age, and area information using the user information. For example, in a case where the user is a twenties female resident in area A, the user group may belong to group A, resident group 20, group 20, and female group. The local information may be selected from the user information, or may be selected by extracting the local keyword included in the currently connected region or the user query.

The user terminal inputs a query such as keyword search, shopping mall visit, game use, URL visit, etc. (508). The related keyword is expanded in the input user query and the scored advertisement content is selectively exposed for the related keyword (510). It can be inferred that the advertising content ranked in the high score is highly effective for the exposed user. Therefore, the advertisement cost of advertisement contents having a high score can be set high. Here, a plurality of exposed advertisement contents may be exposed, the exposed advertisement may be presented in a list format, or may be repeatedly rolled in the same space at regular intervals.

The user query for the exposed advertisement content is fed back to the group profile of the user type to which the user belongs (512). For example, there are cases where a user who inputs a search keyword exposes an advertisement together with a search result, a user clicks an exposed advertisement, or purchases goods or services. This usage pattern is fed back to the profile of the group to which the user belongs, so that the matching value of the keyword and the exposed advertisement is fed back so that the matching accuracy can be improved with time.

6 is a flowchart illustrating an advertisement method according to another embodiment of the present invention.

Referring to FIG. 6, an advertisement server collects a query inputted by a user terminal through an advertisement publisher installed in a user terminal.

In addition, a user group for each gender, age group, and region is defined, a group profile is generated according to a user query in the group, and a related keyword is extracted in a user query in the group (602). The user's query includes search keyword, query item, installation application, purchase product or service, keyword of the edit document, URL address of the site visited by the user, and the like. It is possible to extract keywords from this user query and to extract the related keywords by extending the extracted keywords. The related keywords may be extended from at least one of the keywords in the input same query, the keywords in the continuous query of the same user, and the keywords inputted in advance as the related keywords.

The query related to the related keyword is extracted from the group profile, and the advertisement content is scored for the associated keyword of the user group according to the extracted query and the matching value (604). The user's query related to the related keyword may be extracted from the group profile to determine an advertisement category having a high matching value with the extracted query and the advertisement content may be scored in the advertisement category according to the extracted query and the matching value. This process is advantageous in that the targeted advertisement contents can be scored by reflecting the interest goods or services and the purchase behavior in the query of the user group without determining the advertisement contents according to the related keywords and the matching value.

The advertisement server selects a user group to which the user who is the target of the advertisement exposure belongs. In the present embodiment, the user group selects a user group including at least one of sex, age, and area information of the user using the previous query of the user (606). If the user information is not disclosed, the gender, age, and local information can not be known. Therefore, it is possible to infer the age range, sex, region, etc. of the user by comparing the usage profile in the user's previous query with the group profile stored in the server.

The user terminal inputs a query such as search, visit to a shopping mall, game, or the like (608). In operation 610, the related keyword is expanded in the input user query and the scored advertisement content is selectively exposed for the related keyword. It can be inferred that the advertising content ranked in the high score is highly effective for the exposed user. Therefore, the advertisement cost of advertisement contents having a high score can be set high. Here, a plurality of exposed advertisement contents may be exposed, the exposed advertisement may be presented in a list format, or may be repeatedly rolled in the same space at regular intervals.

The user query for the exposed advertisement content is fed back to the group profile of the user type to which the user belongs (612). For example, there are cases where a user who inputs a search keyword exposes an advertisement together with a search result, a user clicks an exposed advertisement, or purchases goods or services. This usage pattern is fed back to the profile of the group to which the user belongs, so that the matching value of the keyword and the exposed advertisement is fed back so that the matching accuracy can be improved with time.

100: network 110: user terminal
120: Ad server 130: Ad provider
140: Search engine 150:
112: Display 114: Advertiser
116: Ad content 122: Ad selection engine
124: Associated keyword index 126: Advertising content index
128: Profile DB 132: Ad store
142: search index 150a, b, c ...: issued content

Claims (12)

The user terminal storing the publisher;
The advertisement server collecting the query inputted by the user terminal through the advertisement publisher installed in the user terminal;
Wherein the advertisement server classifies the input query into user groups to generate a group profile and extracts a related keyword from the group profile;
The ad server scoring advertisement content according to the association keyword and the matching value in a group profile;
Inputting a query by the user terminal;
The ad server selecting a group profile from the query; And
Wherein the advertisement server extracts a keyword from an input query and exposes the advertisement content scored in the selected group profile to the user terminal through the advertisement publisher for an associated keyword of the extracted keyword,
Wherein scoring the advertisement content comprises:
Extracting a query related to the related keyword in the group profile; And
And scoring the advertisement content for an associated keyword according to the extracted query and the matching value. delete The method according to claim 1,
The step of extracting the related keywords may include:
Extracting keywords from the query; And
Analyzing the keywords to expand at least one of keywords in the same query, keywords in the same user's continuous query, and keywords in the same classified ad category into a related keyword. delete The method according to claim 1,
Wherein the query of the user includes at least one of a search keyword, a product inquiry, a purchase action, an email keyword, an application installation, an edited document keyword, and a URL address of a site visited by a user. The method according to claim 1,
Wherein the selecting of the user group comprises:
And inferring at least one of the user's gender, age range and region using the user's previous query. The method according to claim 1,
Further comprising updating a query of the user terminal for the exposed advertisement content to a group profile of the selected user group A user terminal mounted with an advertisement publisher that transmits an inputted query to an advertisement server;
An advertisement server in communication with the advertisement publisher;
An advertisement storage connected to the advertisement server and storing advertisement contents provided by the advertiser or the advertisement provider;
A related keyword index extracted from a query input to the user terminal;
A group profile in which the input query is classified by user type; And
And an advertisement server for scrolling the advertisement contents stored in the advertisement store according to the association keyword and the matching value in the group profile and exposing the selected advertisement contents to the user terminal,
The advertisement publisher exposes the advertisement selected in the advertisement server to a predetermined area of the display of the user terminal,
The advertisement server includes:
Extracting a query associated with the associated keyword within the group profile,
And scoring the advertisement content with respect to the associated keyword according to the extracted query and the matching value. delete delete The method of claim 8,
The related keyword index is a keyword index,
A keyword in the input query, a keyword in the same query, a keyword in a continuous query of the same user, and a keyword classified into a related keyword in the dictionary. The method of claim 8,
The advertisement server includes:
And updates the group profile by receiving a user query for advertisement contents exposed to the user terminal

Images