JP4728149B2 - Advertisement presentation system and advertisement presentation method - Google Patents

Description

  The present invention relates to an advertisement presenting system and an advertisement presenting method for presenting an optimum advertisement corresponding to the content of any content on the Internet.

  With recent advances in IT (Information Technology) technology, it has become possible to instantly obtain useful information using the Internet, etc., and terminals such as personal computers have become widespread in many homes. is doing. In connection with this, transmission of various information using the Internet is also performed.

  In this information transmission using the Internet, a homepage is established by contracting with a provider, and various HTML (Hyper Text Markup Language) files and image files for transmitting information are stored on the homepage directory, and a browser is stored. By receiving an access request from the information processing apparatus and the like and displaying the contents of the file on the information processing apparatus that is the access request source, it is possible to transmit information to the whole world.

  In addition, with the spread of the Internet, a technique of using the Internet as an advertising medium by pasting a banner advertisement at an arbitrary position on the information provider's home page has become common. The banner advertisement here is an icon that consists of an entry address to the advertiser's homepage and information that visualizes the company name and a simple product copy. The user clicks on this banner advertisement to operate the advertiser. You can enter the homepage and know detailed advertising information (for example, see Patent Document 1).

For this reason, several techniques for providing advertisements corresponding to content have been proposed. For example, a method of expressing the meaning of each content and advertisement on the Internet with a weighted keyword string and providing the advertisement by matching the meanings In addition, a method has been proposed in which the behavior attribute of a visitor of a site is expressed as a weighted keyword string mixed with the meaning of the content, and the advertisement is provided by matching with the keyword corresponding to the advertisement.
JP 2000-194636 A

However, with the method using the former weighted keyword string, the visitors of the site cannot fully understand the meaning of the content and the advertisements provided, and the accuracy and recall of the ads against the needs of the visitors of the site are deteriorated. There is a problem of inviting.
In the latter method, it is difficult to provide advertisements that take into account the combination of site visitor behavioral attributes and content. There is a problem that leads to deterioration.

  Therefore, the present invention has been made in view of the above problems, and provides an appropriate advertisement related to the content of any content on the Internet to meet the needs of the visitors of the site. An object of the present invention is to provide an advertisement presenting system and an advertisement presenting method that prevent the accuracy and reproduction rate of an advertisement from deteriorating.

In order to solve the above problems, the present invention proposes the following matters.
(1) The present invention collects and manages content information management means (for example, the content creator information management unit 10 in FIG. 1) that collects and manages content information from a content creator, and advertisement information from an advertisement provider. Advertising information management means (for example, the advertiser information management section 20 in FIG. 1) and visitor information management means (for example, the visitor information management section 30 in FIG. 1) that collects and manages the visitor behavior information of the site. And the content information received from the content information management means, the advertisement information received from the advertisement information management means, and the behavior information received from the visitor information management means. Based on the data table generation means for generating the table (for example, the advertisement allocation unit 43 in FIG. 1) and the data table, Processing means (for example, the transmission information processing unit 52 in FIG. 1) that generates advertisement information corresponding to content viewed by a site visitor, and the generated advertisement information is embedded in the content to be browsed. An advertisement presenting system characterized by comprising a transmission means for transmitting to a visitor (for example, the transmission / reception unit 51 in FIG. 1 corresponds) is proposed.

  According to the present invention, the content information management means collects and manages the content information from the content creator, and the advertisement information management means collects and manages the advertisement information from the advertisement provider. The visitor information management means collects and manages the visitor behavior information of the site. The data table generation means associates the content information, the advertisement information and the behavior information from the content information received from the content information management means, the advertisement information received from the advertisement information management means, and the behavior information received from the visitor information management means. A table is generated, and the processing means generates advertisement information corresponding to the content viewed by the site visitor based on the data table, and sends the generated advertisement information corresponding to the browsed content to the site visitor. Send. This makes it possible to deliver appropriate advertisements that take into account the combination of site visitor behavioral attributes and content, thus preventing deterioration in the accuracy and recall of advertisements for the needs of site visitors. .

  (2) In the present invention, for (1), the content information management means receives the information related to the content, extracts the site related information of the content from the information related to the content, and generates the content meaning of the graph structure. A content registration means for transmission (for example, the content registration unit 11 in FIG. 1), a content information server for storing site-related information of the transmitted content (for example, the content information server 12 in FIG. 1), A content meaning server for storing the transmitted content meaning (e.g., content meaning server 13 in FIG. 1), and the advertisement information management means receives information related to the advertisement and uses the information to relate to the site of the advertisement. An advertisement registration means that extracts information and generates and transmits the advertisement meaning of the graph structure (example: 1), an advertisement information server (for example, equivalent to the advertisement information server 22 of FIG. 1) for storing the site-related information of the transmitted advertisement, and an advertisement for storing advertisement meanings. A behavior information server (for example, the behavior information server 33 in FIG. 1) that the visitor information management means receives visitor behavior information, and a semantic server (for example, the advertisement meaning server 23 in FIG. 1). Based on the received visitor behavior information, behavior analysis means for analyzing the visitor's behavior pattern and transmitting the behavior meaning and behavior identifier of the graph structure (for example, the behavior analysis unit 31 in FIG. 1 corresponds) And an action meaning server (for example, equivalent to the action meaning server 32 of FIG. 1) for storing the transmitted action meaning and action meaning identifier, and the data table generating means stores the advertisement ontology. Notification ontology server (for example, the advertisement ontology server 42 in FIG. 1), site information related to content from the content information server, content meaning from the content meaning server, and site information related to advertisement from the advertisement information server Based on the advertisement meaning from the advertisement meaning server, the behavior information from the behavior information server, the behavior meaning from the behavior meaning server, and the advertisement ontology from the advertisement ontology server. It is preferable that the advertisement presenting system includes an advertisement assigning unit that generates a data table in which an identifier and an advertisement identifier are associated (for example, the advertisement assigning unit 43 in FIG. 1 corresponds).

  According to this invention, the advertisement allocation means is the site information related to the content from the content information server, the content meaning of the graph structure composed of ontology elements from the content meaning server, the site information related to the advertisement from the advertisement information server, A content identifier based on the advertisement meaning of the graph structure composed of ontology elements from the advertisement meaning server, the action information from the action information server, the action meaning from the action meaning server, and the advertisement ontology from the advertisement ontology server And a data table in which the action meaning identifier and the advertisement identifier are associated with each other. This makes it possible to deliver appropriate advertisements that take into account the combination of site visitor behavioral attributes and content, thus preventing deterioration in the accuracy and recall of advertisements for the needs of site visitors. .

  (3) It is preferable that the present invention is an advertisement presentation system in which the content meaning, advertisement meaning, and action meaning graph structure is configured by ontology elements in (2).

  According to the present invention, the content meaning, the advertising meaning, and the action meaning have a graph structure composed of ontology elements whose elements are conceptualizations of the target world. Can be expressed clearly.

  (4) The present invention refers to the advertisement ontology when receiving information on new content from the content creator or information on new advertisement from the advertisement provider in (2) or (3). Then, semantic analysis means for reconstructing the content meaning of the graph structure composed of ontology elements in the content meaning server and the advertisement meaning of the graph structure composed of ontology elements in the advertisement meaning server (for example, FIG. It is preferable that the advertisement presentation system further includes a semantic analysis unit 41.

  According to the present invention, when the semantic analysis means receives information on new content from the content creator or information on new advertisement from the advertisement provider, the semantic analysis means refers to the advertisement ontology and stores it in the content semantic server. The content meaning of the graph structure composed of ontology elements and the advertisement meaning of the graph structure composed of ontology elements in the advertisement meaning server are reconstructed. Therefore, even when information about new content or information about advertisements is generated, the content meaning and advertisement meaning corresponding to the information is constructed, and based on these, the advertisement distribution corresponding to the needs of the visitor of the site is made. It can be performed.

  (5) The present invention receives information related to content, extracts site related information of the content from the information related to the content, generates content meaning of a graph structure, and relates the content meaning to the site related to the extracted content The first step of storing information (for example, corresponding to step S101 in FIG. 2), information related to the advertisement is received, the site related information of the advertisement is extracted from the information, and the advertisement meaning of the graph structure is generated. A second step of storing the advertisement meaning and the site-related information of the extracted advertisement (for example, step S102 in FIG. 2 corresponds), receiving the visitor behavior information, and receiving the received visitor behavior information Based on the behavior pattern of the visitor, and transmits the behavior meaning and behavior information of the graph structure, and the transmitted behavior meaning and behavior information. 3 (for example, corresponding to step S103 in FIG. 2), a prestored advertisement ontology, site information related to the content, the content meaning, site information related to the advertisement, and the advertisement A fourth step (for example, corresponding to step S104 in FIG. 2) of generating a data table in which the content identifier, the behavior meaning identifier, and the advertisement identifier are associated with each other based on the meaning, the behavior information, and the behavior meaning; And a fifth step (for example, corresponding to step S105 in FIG. 2) of transmitting advertisement information corresponding to the content viewed by the site visitor to the site visitor based on the data table. We have proposed an advertisement presentation method characterized by this.

  According to the present invention, first, content-related information is received, site-related information of content is extracted from the information about content, content meaning of a graph structure is generated, content meaning, and site-related information of the extracted content are Is stored. Also, information related to the advertisement is received, site related information of the advertisement is extracted from the information, advertisement meaning of the graph structure is generated, and the advertisement meaning and the site related information of the extracted advertisement are stored. Furthermore, it receives the visitor behavior information, analyzes the visitor behavior pattern based on the received visitor behavior information, transmits the behavior meaning and behavior information of the graph structure, and the transmitted behavior meaning and behavior information. And store. And based on the pre-stored advertisement ontology, content site information, content meaning, advertisement site information, advertisement meaning, behavior information, and behavior meaning, content identifier, behavior meaning identifier, and advertisement identifier Is generated, and based on the generated data table, appropriate advertisement information corresponding to the content viewed by the site visitor is transmitted to the site visitor. This makes it possible to deliver appropriate advertisements that take into account the combination of site visitor behavioral attributes and content, thus preventing deterioration in the accuracy and recall of advertisements for the needs of site visitors. .

  (6) Moreover, it is preferable that the present invention is an advertising method characterized in that the graph structure of the content meaning, the advertising meaning, and the action meaning is configured by ontology elements with respect to (5).

  According to the present invention, the content meaning, the advertising meaning, and the action meaning have a graph structure composed of ontology elements whose elements are conceptualizations of the target world. Can be expressed clearly.

  (7) Further, according to (5) or (6), the present invention refers to the pre-stored advertisement ontology when the information about new content or information about new advertisement is received, and the content meaning It is preferable that the advertisement presentation method further includes a step of reconstructing the advertisement meaning.

  According to the present invention, when information related to new content or information related to a new advertisement is received, the content meaning and the advertisement meaning are reconstructed by referring to the previously stored advertisement ontology. Therefore, even when information about new content or information about advertisements is generated, the content meaning and advertisement meaning corresponding to the information is constructed, and based on these, the advertisement distribution corresponding to the needs of the visitor of the site is made. It can be performed.

  According to the present invention, for an arbitrary content on the Internet, an appropriate advertisement related to the content is provided to prevent deterioration in the accuracy and reproduction rate of the advertisement for the needs of the site visitors. There is an effect that can be.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
Note that the components in the present embodiment can be appropriately replaced with existing components and the like, and various variations including combinations with other existing components are possible. The description does not limit the contents of the invention described in the claims.

  As shown in FIG. 3, the advertisement presentation system according to the present embodiment includes a content meaning of a graph structure configured by ontology elements generated from content information and a graph structure configured by ontology elements generated from advertisement information. In the content viewed by the site visitor, the site visitor's content is related to the advertisement ontology and the behavioral meaning of the graph structure composed of ontology elements generated from the site visitor's behavior information. Deliver appropriate ads that meet your needs.

  Therefore, before explaining the details of the configuration and processing operation of the advertisement presentation system according to the present embodiment, the content meaning, advertisement meaning, and action meaning of the graph structure composed of the advertisement ontology and ontology elements will be described.

<Advertising Ontology>
The advertisement ontology will be described with reference to FIG.
In general, an ontology is a system of concepts, and the underlying concept is conceptualization. Conceptualization refers to concepts that exist in the target world and the relationships that are established between them. Therefore, the ontology means a graph structure as shown in FIG. 3 having the conceptualization of the target world as an element.

  The advertisement ontology in the present embodiment includes the classification of products and services that are the concept of advertisement, the behavioral purpose and temporal transition that is the concept of relation, abstraction, etc. This is a graph structure showing general concepts in natural language, etc.

  The ontology is represented by a concept definition table as shown in FIG. 4A and an inter-concept relationship definition table as shown in FIG. The concept definition table is composed of concept identifiers and concept contents as shown in FIG. 4A, and the inter-concept relationship definition table is shown in FIG. It consists of a concept identifier and a concept type.

  In the example of FIG. 4B, for example, the original concept identifier “120” corresponds to the concept content “food” in the concept definition table, and the previous concept identifier “12” corresponds to the concept content “pasta” in the concept definition table. Correspond. In this example, since “pasta” has a relationship of embodying “food”, the concept type between the two is “materialization”.

  As another example, the original concept identifier “250” corresponds to the concept content “car” in the concept definition table, and the previous concept identifier “275” corresponds to the concept content “tire” in the concept definition table. In this case, since the “tire” is one component of the “car”, the concept type between the two is “component”. That is, the advertisement ontology in the present embodiment shows such a relationship in a graph structure with respect to the advertisement.

<Content meaning / Advertising meaning / Action meaning>
Next, content meaning, advertisement meaning, and action meaning will be described with reference to FIG.
The content meaning is a representation of the subject of the content in a graph composed of ontology elements, the advertising meaning is a representation of the intent of the advertisement in a graph composed of ontology elements, and the action meaning is a visitor It expresses the behavioral intentions and interests in a graph composed of ontology elements.

  Here, the graph is based on a semantic graph notation table as shown in FIG. 5, and the content is illustrated in FIG. 5, but the same applies to the advertisement meaning and the action meaning. As shown in FIG. 5, the semantic graph notation table is composed of an element identifier, a display label, a concept identifier list, an original element identifier, and a destination element identifier. The term extracted from the content information is included in the display label. It is written.

  As shown in the column of the element identifier “9” in FIG. 5, the elements in which the original element identifier and the destination element identifier are defined (in FIG. 5, the “pasta” of the element identifier “1” and the element identifier “2” “Boiled”) are connected by arcs (solid arrows) shown in FIG. 3, and undefined elements are expressed as nodes (parts shown by squares) shown in FIG. In addition, when the identifiers described in the concept identifier list and the concept identifiers in the concept definition table shown in FIG. 4 match, as shown in FIG. 3, a link is established between the content meaning and the advertisement ontology to associate them. (Dotted line in FIG. 3).

  Further, as shown in FIG. 3, each semantic notation has a graph structure composed of nodes indicated by squares and arcs indicated by solid arrows in the figure, and each node and arc is represented by It can have an advertisement ontology element concept and a plurality of links (broken lines in the figure) indicated by square diagonal lines. In addition, when generating content meaning, advertising meaning, and action meaning of graph structure composed of ontology elements from content information, advertisement information, and action information, by using a method using machine learning and dictionary and rules, It can be executed automatically.

<Advertising system configuration>
Based on the above contents, the configuration of the advertisement presentation system according to the present embodiment will be described.
As shown in FIG. 1, the advertisement presentation system according to the present embodiment includes a content creator information management unit 10, an advertiser information management unit 20, a visitor information management unit 30, an analysis processing unit 40, and a visitor. The front processing unit 50 is directed to.

  The content creator information management unit 10 includes a content registration unit 11, a content information server 12, and a content meaning server 13. The advertiser information management unit 20 includes an advertisement registration unit 21 and an advertisement information server 22. The visitor information management unit 30 is composed of a behavior analysis unit 31, a behavior meaning server 32, and a behavior information server 33. The analysis processing unit 40 is composed of a semantic analysis unit 41. The advertisement ontology server 42, the advertisement allocation unit 43, and the data table storage server 44, and the visitor front processing unit 50 includes a transmission / reception unit 51 and a transmission information processing unit 52.

  The content creator information management unit 10 receives information about the content from the content creator, and transmits the content information and the content meaning to the content information server 12 and the content meaning server 13, respectively.

  Specifically, the content registration unit 11 receives the URL and keywords related to the content from the content creator, the site category, the rival advertisement designation, the content range designation information, etc., and directly accesses the URL of the received data, Site-related information, such as links, estimated templates, estimated categories, natural language mark, page rank, etc., to collect content information and generate content meaning of graph structure composed of ontology elements The extracted content information is stored in the content information server 12 and the generated content meaning is stored in the content meaning server 13. In addition, the content meaning and site related information of the graph structure comprised of ontology elements may be directly acquired from the content creator, and these data may be stored in the content information server 12 or the content meaning server 13.

  The advertiser information management unit 20 receives information related to the advertisement from the advertiser, and transmits the advertisement information and the advertisement meaning to the advertisement information server 22 and the advertisement meaning server 23, respectively.

  Specifically, the advertisement registration unit 21 receives an advertisement URL, a keyword category, a user image designation (for example, information on a heavy net user, a man in their 30s, a mobile user, etc.), an advertisement, etc. , By directly accessing the URL of the received data, collecting information related to the site of the advertisement, such as links, estimated templates, estimated categories, natural language mark, page rank, etc. The advertisement meaning of the graph structure constituted by the above is generated, the extracted advertisement information is stored in the advertisement information server 22, and the generated advertisement meaning is stored in the advertisement meaning server 23. It is also possible to directly acquire the advertisement meaning and site related information of the graph structure composed of ontology elements from the advertiser and store these data in the advertisement information server 22 and the advertisement meaning server 23.

  The visitor information management unit 30 receives the visitor's action information, and transmits the action meaning of the graph structure composed of ontology elements and an action meaning identifier described later to the action meaning server 32. The behavior information received from the visitor is stored in the behavior information server.

  Specifically, from the transmitting / receiving unit 51 in the front processing unit 50 for visitors, referrer, cookie, application information by an RSS (Rich Site Summary) reader, device information such as a mobile phone or PDA (Personal Data Assistants), GPS ( Location information, time information, IP (Internet Protocol) address, advertisement click information, movement history between pages on which advertisement HTML is placed, personal information such as age and gender, etc. are sent via the action information server 33. The behavior analysis unit 31 receives and analyzes the pattern of the visitor behavior information based on the received information, and the behavior meaning of the graph structure composed of ontology elements and the behavior meaning for accessing this behavior meaning To generate a Besshi.

  The analysis processing unit 40 is generated by the content information and content meaning generated by the content creator information management unit 10 and the advertisement information and advertisement meaning generated by the advertiser information management unit 20 and by the visitor information management unit 30. The action information, action meaning, and advertisement ontology stored in the advertisement ontology server are received, and a data table that defines the relationship between the content, action, and advertisement is stored in the data table storage server 44.

  Here, the advertising ontology includes business domains, product classifications, company names, region names, behavior types, visitor attributes, inter-concept relationships, etc. as conceptual items, and each conceptual item has an alias notation list. . In addition, inter-concept relationships such as hierarchical relationships (is_a), inclusion relationships (part_of), objectives (objects), and methods (methods) are freely defined between concept items, and are stored in the advertisement ontology server 42 in advance. .

  Specifically, the processing of the analysis processing unit 40 is performed by the advertisement allocation unit 43 receiving the content information, content meaning, advertising information, advertising meaning, behavior information, behavior meaning, and advertisement ontology, and an advertisement similar to the received content meaning. Search the meaning and send a data table that defines the relationship between content, behavior and advertisement.

  In similarity determination, similarity determination conditions may be adjusted using word weighting. Further, the abstract concrete level of the concept may be adjusted using a relationship such as a vertical relationship (is_a) or an inclusion relationship (part_of) between the advertising ontology concept items. In addition, replacement with related concept items may be performed using a relationship such as an object or method between advertisement ontology concept items. Furthermore, you may switch the related conceptual item according to a visitor's action type using action meaning.

  The semantic analysis unit 41 in the analysis processing unit 40 receives new content information and advertisement information from the content information server 12 or the advertisement information server 22 and an advertisement ontology from the advertisement ontology server, and is composed of ontology elements. The content meaning or advertisement meaning of the graph structure is reconstructed and transmitted to the content meaning server 13 or the advertisement meaning server 23.

  Specifically, natural language data is subjected to morphological element analysis, cut out in units of words, and converted into content meaning or advertisement meaning based on this result and received information. In addition, when the content meaning or advertisement meaning of the graph structure comprised by ontology element is provided from the content creator or the advertiser, these may be merged. In the above description, the extraction in units of words by morphological element analysis has been exemplified, but further syntax analysis may be performed, or weighting may be performed using word statistical information.

  The visitor front processing unit 50 receives a browsing request for a specific site from a visitor of the site and information related to the visitor, and creates a data table that defines the relationship between content, action, and advertisement from the data table storage server 44. Receives and distributes the requested content to the site visitors, and distributes advertisements corresponding to the needs of the site visitors.

  Specifically, the transmission / reception unit 51 receives a site browsing request from a site visitor and information about the visitor. The transmission information processing unit 52 acquires the content from the content information server 12 by accessing the URL of the site requested to be browsed. Further, the transmission / reception unit 51 transmits the visitor's behavior information to the visitor information management unit 30, and the transmission information processing unit 52 acquires the visitor's behavior meaning identifier from the behavior meaning server 32.

  Further, the transmission information processing unit 52 acquires a data table from the data table storage server 44, and acquires an identifier of an advertisement to be posted on the content using the URL of the site and the visitor's action meaning identifier. The transmission information processing unit 52 also acquires an advertisement from the advertisement information server 22 using the acquired advertisement identifier, embeds the advertisement in the content to be distributed, and transmits the advertisement to the transmission / reception unit 51. The transmission / reception unit 51 distributes the received content to visitors.

<Advertising example>
Next, the advertisement allocation will be specifically described with reference to FIGS.
For example, a case as shown in FIG. 8 is assumed. Here, the contents accessed by the visitors A and B of the site are the name “wisdom bag” and the content ID “592456”, and the meaning of this content is “pasta → boil”.

  Further, there are “pasta food store advertisement” (advertisement ID = ad1), “recipe main advertisement” (advertisement ID = ad2), “restaurant advertisement” (advertisement ID = ad3) as advertisement information. The advertisement meanings are “pasta → cooking” and “food ingredients → purchase”, the advertisement meaning of “recipe main advertisement” is “food ingredients → cooking”, and the advertisement meaning of “restaurant advertisement” is “food ingredients → eat”.

  In addition, the behavior information of visitor A is “reaching content via search engine at noon”, the meaning of this behavior is “single daytime”, and the behavior information of visitor B is “every day, at night” “Reach content after browsing the recipe site”, and the action meaning is “continuous at night, target; recipe”. Furthermore, on the advertising ontology, there is a relationship of “pasta → food” and “boiled → cooking”.

  In the above case, assuming that the visitor A visits the content “wisdom bag, content ID = 592456”, first, as shown in FIG. 7, the content meaning generated from the content in the advertisement allocation unit 43 ("Pasta → boil") is expanded according to the abstraction rule to obtain "pasta → boil", "foodstuff → boil", "pasta → cooking", and "foodstuff → cooking". On the other hand, as shown in FIG. 7, advertisement meanings (“pasta food store advertisement”, “recipe book advertisement”, “restaurant advertisement”) are developed according to abstraction rules, and “pasta → cooking”, “food ingredients → purchase”. , “Food → Cooking”, “Food → Eat” By matching both of these, a “pasta food store advertisement” (advertisement ID = ad1) and a “recipe main advertisement” (advertisement ID = ad2) are obtained.

  Next, in order to narrow down the options obtained by the matching in the advertisement allocation unit 43, the behavior meanings of the visitors A and B, that is, for the visitor A, “single lunch → eat”, “lunch single purchase → purchase” ", For visitor B, use" continuous at night, subject; recipe-> cooking ".

  As a result, “pasta food store advertisement” (advertisement ID = ad1) is optimal as an advertisement distributed to the visitor A together with the content, and “pasta” The “food shop advertisement” (advertisement ID = ad1) and the “recipe main advertisement” (advertisement ID = ad2) are optimal.

<Processing of advertisement presentation system>
Next, the processing operation of the advertisement presentation system will be described with reference to FIG.
The content registration unit 11 receives content related information such as URLs and keywords related to content from content creators, site categories, rival advertisement designation, content range designation information, and the like, and extracts content information extracted from the received content related information. In the content information server 12, the content meaning of the graph structure composed of ontology elements generated from the content related information is stored in the content meaning server 13, and the stored content information and content meaning are transmitted to the advertisement allocation unit 43 ( Step S101).

  The advertisement registration unit 21 receives an advertisement URL, a keyword / category, a user image designation (for example, information on a heavy net user, a man in their 30s, a mobile user, etc.), a promotional sentence, etc., and a received advertisement. The advertisement information extracted from the related information is stored in the advertisement information server 22, the advertisement meaning of the graph structure composed of ontology elements generated from the advertisement related information is stored in the advertisement meaning server 23, and the stored advertisement information and advertisement meaning are respectively advertised It transmits to the allocation part 43 (step S102).

  The behavior analysis unit 31 receives information from the sending / receiving unit 51 in the front processing unit 50 for visitors from the referrer, cookie, application information by an RSS reader, device information such as a mobile phone or PDA, position information by GPS, time information, IP address , Advertisement click information, movement history between pages with advertisement HTML, personal information such as age and gender etc. are received via the behavior information server 33, and the received visitor behavior information pattern is analyzed to The meaning is generated, the generated action meaning is stored in the action meaning server 32, and the stored action information and action meaning are transmitted to the advertisement assigning unit 43 (step S103).

  Based on the received content information, content meaning, advertising information, advertising meaning, behavior information, behavior meaning, and advertising ontology, the advertisement allocation unit 43 generates a data table in which the content, behavior, and advertisement are associated with each other, and a data table storage server 44 (step S104).

  The transmission / reception unit 51 receives a site browsing request from a visitor, and the transmission information processing unit 52 is optimized based on the site browsing request received by the transmission / reception unit 51 and the data table received from the data table storage server 44. The advertisement is embedded in the content and transmitted to the visitor (step S105).

  Therefore, according to this embodiment, the content meaning of the graph structure composed of ontology elements generated from the content information, the advertisement meaning of the graph structure composed of ontology elements generated from the advertisement information, and the site visit By selecting the advertisement corresponding to the content by associating the behavior meaning of the graph structure composed of ontology elements generated from the behavior information of the user with the advertisement ontology and selecting the advertisement corresponding to the content, Appropriate advertisements that meet the needs of visitors can be distributed.

  As described above, the embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes a design and the like within a scope not departing from the gist of the present invention. .

It is a block diagram of the advertisement presentation system which concerns on this embodiment. It is a processing flow of the advertisement presentation system which concerns on this embodiment. It is the figure which showed the relationship between content meaning, advertisement meaning, action meaning, and advertisement ontology. It is a figure which illustrates a concept definition table and a concept relationship definition table. It is the figure which illustrated the semantic graph description table. It is the figure which illustrated the data table which showed the relationship between a content, action, and an advertisement. It is the figure which illustrated matching processing. It is the figure which illustrated the allocation example of advertisement, and the allocation logic.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Content producer information management part (content information management means), 11 ... Content registration part (content registration means), 12 ... Content information server (content information server), 13 ... Content meaning server (Content meaning server), 20 ... advertiser information management unit (advertising information management means), 21 ... advertisement registration unit (advertisement registration means), 22 ... advertisement information server (advertisement information server), 23. ..Advertising meaning server (advertising meaning server), 30 ... Visitor information management section (visitor information management means), 31 ... Behavior analysis section (behavior analysis means), 32 ... Behavior meaning server (behavior) Meaning server), 33 ... behavior information server (behavior information server), 40 ... analysis processing unit (processing means), 41 ... semantic analysis unit (semantic analysis means 9, 42 ... advertisement on) Logistic server (advertising ontology server), 43... Advertisement allocation unit (data table generation means, advertisement allocation means), 44... Data table storage server, 50... Front processing unit for visitors, 51. Transmission / reception unit (transmission means), 52... Transmission information processing unit

Claims (7)

Content information management means for collecting and managing content information from content creators;
Advertising information management means for collecting and managing advertising information from advertising providers,
A visitor information management means for collecting and managing visitor behavior information on the site,
A content meaning that is a graph structure generated by extracting a plurality of terms that are ontology elements from the content information, an advertisement meaning that is a graph structure generated by extracting a plurality of terms that are ontology elements from the advertisement information, and the behavior the term is ontology elements from the action means is a graph structure which is generated by a plurality extracted from the information, the data table generating means for generating a data table that associates the content information, the advertisement information and the action information,
Processing means for generating advertisement information corresponding to the content viewed by the visitor of the site by narrowing down the options obtained by matching the content meaning and the advertisement meaning based on the data table by the action meaning ; ,
A transmitting means for embedding the generated advertisement information in the content to be browsed and transmitting it to a visitor of the site;
An advertisement presentation system comprising: The content information management means receives the information related to the contents, extracts the site related information of the contents from the information related to the contents, and generates and transmits the content meaning of the graph structure, and the transmitted contents A content information server for storing the site related information, and a content meaning server for storing the transmitted content meaning,
The advertisement information management means receives advertisement-related information, extracts advertisement-related site information from the information, generates and transmits an advertisement meaning in a graph structure, and transmits the advertisement advertisement site. An advertisement information server for storing related information; and an advertisement meaning server for storing advertisement meaning,
The visitor information management means receives visitor behavior information, analyzes the visitor's behavior pattern based on the received visitor behavior information, and obtains the behavior meaning and behavior identifier of the graph structure. Action analysis means for transmitting, and an action meaning server for storing the transmitted action meaning and action meaning identifier,
An advertisement ontology server in which the data table generation means stores an advertisement ontology; site information relating to content from the content information server; content meaning from the content meaning server; site information relating to advertisement from the advertisement information server; Based on the advertisement meaning from the advertisement meaning server, the behavior information from the behavior information server, the behavior meaning from the behavior meaning server, and the advertisement ontology from the advertisement ontology server, The advertisement presenting system according to claim 1, further comprising an advertisement assigning unit that generates a data table in which the advertisement identifier is associated with the advertisement identifier.   The advertisement presentation system according to claim 2, wherein the graph structure of the content meaning, the advertisement meaning, and the action meaning is configured by ontology elements.   A graph composed of ontology elements in the content meaning server with reference to the advertisement ontology when information about new content from the content creator or information about a new advertisement from the advertisement provider is received. 4. The advertisement presentation system according to claim 2, further comprising semantic analysis means for reconstructing the advertisement meaning of the graph structure composed of the content meaning of the structure and the ontology element in the advertisement meaning server. Receiving information related to content, extracting site-related information of content from the information related to content, generating content meaning in a graph structure, and storing the content meaning and site-related information of the extracted content; And the steps
A second step of receiving information related to the advertisement, extracting advertisement-related information of the advertisement from the information, generating an advertisement meaning of the graph structure, and storing the advertisement meaning and the extracted advertisement-related site-related information When,
Receives visitor behavior information, analyzes visitor behavior patterns based on the received visitor behavior information, transmits behavior meaning and behavior information of the graph structure, and transmits the transmitted behavior meaning and behavior A third step of storing information;
Content identifier and action meaning based on pre-stored advertisement ontology, site information related to the content, content meaning, site information related to the advertisement, the advertisement meaning, the action information, and the action meaning A fourth step of generating a data table associating the identifier and the advertisement identifier;
Based on the data table, by narrowing down the options obtained by matching the content meaning and the advertisement meaning by the action meaning, advertisement information corresponding to the content viewed by the site visitor is sent to the site visitor. A fifth step of transmitting;
An advertisement presentation method characterized by comprising:   The advertisement presentation method according to claim 5, wherein the graph structure of the content meaning, the advertisement meaning, and the action meaning is configured by ontology elements.   The method further comprises the step of reconstructing the content meaning and the advertisement meaning by referring to the prestored advertisement ontology when information on new content or information on a new advertisement is received. Item 7. The advertisement presenting method according to Item 5 or 6.

Images