JP6362577B2 - Information processing apparatus and display article selection system - Google Patents

Description

  The present invention relates to an information processing apparatus and a display article selection system, and more particularly, to a technique for easily touching a user's eyes with contents including information that an advertiser wants to recognize in accordance with the interest of the user.

  In general advertisements, advertisements with the same content and format are widely presented to media users. For this reason, a user will see the same advertisement repeatedly, and the value as information will fall rapidly and it will only become annoying.

  Advertisements using the Internet as a media have been devised in various ways to increase their appeal. For example, there is a style in which an advertiser pays an advertisement fee to a web magazine or the like and requests publication of an article advertisement as a contrivance for advertisement distribution and marketing. A well-designed article advertisement can catch the eyes of many people.

  However, article advertisements must be formatted for each web magazine, and are expensive to create. Also, a completely different article and a different article advertisement that the user recognizes cost even more. Therefore, in fact, even in different web magazines, the number of article advertisements with the same contents that differ only in appearance will increase, the user will see the same thing repeatedly, the feeling of advertisement will increase, the feeling of articles will decrease, as with normal advertisements The value of information goes down.

  On the other hand, in the technical aspect, for example, there is a device for providing an advertisement from an advertisement network using a third party cookie. First, an advertising space is embedded in a web page that the website provides for browsing by the user. When the user browses the web page, the advertisement network server is accessed from the advertisement space, and the advertisement is displayed. Since the user profile can be specified by using a third party cookie, the advertising network can perform targeting (for example, Patent Document 1).

  By targeting, an advertisement useful for the user can be presented (the user's troublesomeness can be reduced). From the perspective of advertisers who pay advertising fees for the number of advertisements viewed, the effectiveness of advertising increases, so it can be said that it is efficient in terms of both cost and effectiveness.

  However, the method of delivering advertisements from the advertisement network has a problem in the accuracy of targeting because accurate name identification is difficult from the viewpoint of privacy protection. Even if targeting is possible, there is a problem that the advertisement insertion rate is low because an article inventory with a high cost compared with a normal advertisement has a small inventory. For this reason, article advertisements are generally not targeted or performed loosely and are often presented to all media users, which is bothersome for the user and has a low effect on the browsing number ratio.

Special table 2007-507801 gazette

  In any of the above cases, the advertisement is troublesome for the user. No matter how useful the advertisements are, there is little advertisement inventory, and the main reason is that the same advertisements appear repeatedly as the advertisements that are useful for the user are selected carefully.

  The present invention has been made in view of the above circumstances, and it is an object of the present invention to make it easy for the user's eyes to touch content including information that the advertiser wants to recognize in line with the user's interests.

  The first apparatus of the present invention that achieves the above object includes a selection unit that selects one or more articles from an article group that are smaller than the number of articles in the article group, and an article selection condition that is advertisement information. Selection condition input means for inputting to the display, and display means for displaying the article selected by the selection means. The selection means selects an article that matches the article selection condition input by the selection condition input means. The display article is selected probabilistically after increasing the probability of being displayed.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to make a user's eyes easy to touch the content containing the information which an advertiser wants to recognize according to a user's interest.

It is a block diagram which shows the function structure of 1st Embodiment. It is a block diagram which shows the structure of the system of 2nd Embodiment. It is a block diagram which shows the function structure of the system of 2nd Embodiment. It is a sequence diagram which shows the flow of a process of 2nd Embodiment. It is a figure which shows an example of the display output in 2nd Embodiment. It is a block diagram which shows the structure of the system of 3rd Embodiment. It is a block diagram which shows the function structure of the system of 3rd Embodiment. It is a figure which shows an example of the display output in 3rd Embodiment. It is a block diagram which shows the function structure of the system of 4th Embodiment.

<First Embodiment>
Hereinafter, a first embodiment will be described with reference to the drawings.
FIG. 1 shows a functional configuration of the present embodiment. Each block shown in the figure is a logical functional block generated by causing a computer system to execute a software program. As shown in the figure, the present embodiment includes a selection unit 1, a display unit 2, and a selection condition input unit 3. Each means may be executed on a processor of a physically different computer. In the following, in order to simplify the description, it is described as information processing performed on a single computer, but it goes without saying that the embodiment is not limited to this.

  The selection means 1 performs information processing for selecting (pickup) some of one to a plurality of article groups. This selection is preferably a selection for displaying the articles in order at the position where the user's attention of the present embodiment gathers in the display means 2 in the subsequent stage.

  The display means 2 performs information processing for determining the layout of the article selected by the selection means 1 and displaying it on a liquid crystal display or the like. This information processing includes, for example, a mode in which the web server determines the layout of the web page and displays it on the web client (so-called browser).

  The selection condition input unit 3 performs information processing for inputting the article selection condition used when the selection unit 1 performs information processing related to article selection to the selection unit 1. A specific aspect of this includes an aspect in which an article selection condition is received from an external server via a network and is passed to the selection means 1. In addition, a mode is also included in which information sources such as web page groups on the Internet are searched with a predetermined keyword or the like, and the obtained search results are input to the selection means 1 as article selection conditions.

  The selection means 1 selects an article based on the probability set for each article so as to meet the user's interest. Although not limited, for example, a score is set for each article, and the score of an article in the sum of the scores is set as an article from which the article is selected. Here, with respect to an article that matches the article selection condition input from the selection condition input means 3, the selection means 1 performs a probabilistic article selection after increasing the probability that the article is selected. Although the specific aspect of the information processing which raises a probability is not limited, For example, raising the above-mentioned score etc. are mentioned.

  Here, the article selection condition input to the selection means 1 by the selection condition input means 3 is preferably a condition related to information that the advertiser wants to recognize. Therefore, the selection means 1 makes a probability selection from an article group after increasing the probability that an article that matches an article selection condition determined based on advertisement information will be selected. The articles selected in this way are preferably displayed by the display means 2 in order of the user's attention.

  According to the present embodiment, an article having content that meets the user's interest and that includes information that the advertiser wants to recognize is presented in a form that is easy to touch the user's eyes.

  The operational effect of the present embodiment will be described by taking a portal site displaying a list of pickup articles as an example. Conventionally, a pickup article is selected based on the number of citations such as access ranking and Twitter (registered trademark). If the article contains information that the user wants to recognize and may be selected, the possibility of picking up is increased. Individual articles are not necessarily picked up, but the chances are high, and the probability of the display of a group of articles containing information that the advertiser wants to recognize can be reduced while reducing the chance that the user will see the same article repeatedly. The number of users who have seen advertisement articles is expected to increase even when the user group is regarded as a mass.

  The example of the portal site is merely an example, and hereinafter, embodiments more specific to this embodiment will be shown as second to fourth embodiments.

<Second Embodiment>
Hereinafter, the second embodiment will be described with reference to the drawings.

FIG. 2 shows the system configuration of this embodiment.
As shown in the figure, the computer system according to the present embodiment is composed of a server / client system. The computer system includes a client 10, an evaluation server 11, an article distribution server 12, and an advertisement information distribution server 14.

  Although not limited, the evaluation server 11 is preferably a server managed by a business operator (advertiser) who places an advertisement. The client 10 is preferably an end-user personal device (such as a laptop computer, a smart phone, or a smart pad) as an example. There are a plurality of clients 10.

  The article distribution server 12 is a server that distributes articles to clients. An article here means a group of things written in natural language. The article includes, for example, one news article distributed on the Internet web and one entry of a web log. A specific example of the article distribution server 12 is an HTTP server. Article distribution may be a push type or a pull type. In this embodiment, the client 10 pulls an article from the article distribution server 12. There are a plurality of article distribution servers 12.

  The advertisement information distribution server 14 is a server that distributes to the client 10 and inputs “specific keywords” related to the article selection condition at the time of selecting an article in the client 10.

In FIG. 3, the function structure of the principal part of this embodiment is shown.
As illustrated, the client 10 includes an advertisement article search unit 101, a social importance level addition unit 102, a social importance level calculation unit 103, an individual interest level calculation unit 104, a list display article selection unit 105, and an advertisement article list display count. Reporting means 106 and display output means 107 are provided. Further, the evaluation server 11 includes an advertisement article list display count counting unit 108. The configuration of the plurality of clients 10 is the same.

  In summary, the client 10 acquires articles distributed by the article distribution server 12, selects some of them, and performs information processing for display output on the display output means 107. Among them, the importance is weighted to the articles extracted by the search by the specific keyword. In the following, functions of each means shown in the drawings for performing these information processes will be described.

  The advertisement article search means 101 has a function of searching for articles (such as articles and press articles whose press releases are taken up) that match a specific keyword designated by the advertiser. Here, a company name or a product name is assumed as the specific keyword. The “article” described here is the same as the article distributed by the article distribution server 12, but as an example, an article created by a news distribution site requesting a professional writer (hereinafter referred to as “advertisement article”). In addition, it also includes entry articles posted on weblogs established by amateur individuals. It also includes entry articles posted on short posting sites such as Twitter (registered trademark).

  Further, “searching for articles” means information processing for listing the articles related to the specific keyword according to at least the strength of the relation. In order to perform this information processing, the advertisement article search means 101 may use an external search engine or library.

  Further, the advertisement article search means 101 preferably has a function of determining whether an article writer gives a positive evaluation or a negative evaluation for a specific keyword in the searched article. . For this purpose, a conventionally known method can be used. For example, a method can be used in which a positive / negative adjective dictionary is prepared and an adjective that modifies a specific keyword is included in which adjective dictionary.

  Moreover, it is not always the case that a positive determination is made only for a keyword that represents an advertiser's product name, and a negative determination may be made for a keyword that represents a problem to be solved by the product. For example, a sentence such as “Laundry detergent that can shorten useless rinsing time” is not a word that modifies the product name “NEC”, but “useless” that modifies the problem “Rinse time” that the product solves. It can be considered that the product is favorably evaluated by the negative word.

  The specific keyword is acquired and input by the advertisement article search means 101 from the advertisement information distribution server 14. For example, it is preferable that the advertisement article search unit 101 periodically acquires a specific keyword and periodically updates the keyword list.

  The social importance adding means 102 has a function of adding a positive offset to the social importance to the articles searched and listed by the advertisement article searching means 101. In the present embodiment, a parameter “social importance” is set for each article. The more social importance, the more important the article is for many different segments.

  The social importance calculating unit 103 has a function of calculating the social importance of a plurality of articles distributed from the article distribution server 12 including the articles to which the offset is given by the social importance adding unit 102. The social importance calculation method is not particularly limited. For example, the social importance calculation method may be calculated according to the number of article citations or the number of trackbacks, or using the importance set by the article distribution server 12. Also good. Alternatively, it may be calculated using a trendy search keyword published by an external ranking site.

  The personal interest degree calculation means 104 is a list display article of the distribution articles acquired from the article distribution server 12 by the client 10 in accordance with the personal hobby preference of the user of the client 10 and the interest in the occasion. A function for calculating a parameter (individual interest) that affects the ease of selection in the selection means 105 is provided.

  For example, the distribution articles can be classified into genres such as “politics”, “economy”, and “sports”. If the user is interested in a certain genre, the ease of selecting a distribution article belonging to the genre is increased. The personal interest level is a parameter set for each distribution article.

  The list display article selection unit 105 outputs the social importance degree calculated by the social importance degree calculation unit 103 and the distribution article (plurality) calculated by the personal interest degree calculation unit 104 to the display output unit 107. A function for selecting a distribution article to be displayed and output is provided. Selected distribution articles are displayed in a list.

  The display output unit 107 constitutes a display screen of the client 10 and has a function of displaying and outputting the distribution article selected by the list display article selection unit 105.

  The advertisement article list display frequency reporting means 106 is the number of times the article extracted by the advertisement article search means 101 using a specific keyword is selected by the list display article selection means 105 and displayed on the display output means 107 (“ A function of reporting the number of times of list display)) to the external evaluation server 11 is provided. The list display count may be the number selected by the list display article selection means 105. Further, when there are a plurality of articles searched and extracted by the advertisement article search means 101 using a specific keyword, the articles are aggregated and reported for each article.

  The evaluation server 11 includes an advertisement article list display count counting unit 108. The advertisement article list display count counting means 108 has a function of counting the list display counts reported by the advertisement article list display count reporting means 106 of the client 10. As described above, a plurality of clients 10 exist in the system. By receiving reports of the number of times of list display from each of them and counting them, it is possible to measure the advertising effect of inputting a specific keyword to each client.

FIG. 4 shows a processing flow of the present embodiment.
In the present embodiment, as shown in FIG. 4, first, the client 10 inquires of the article distribution server 12 about whether or not the distribution article is updated (S101). The article distribution server 12 distributes articles accordingly (S102).

  Next, the client 10 inquires whether there is an update of the “specific keyword” (S103). This timing may be configured to be performed periodically, for example, once a month or once a week. In the present embodiment, it is performed after the acquisition of the distribution article. On the other hand, the evaluation server 11 returns “specific keyword” to the client 10 as a response (S104).

  The client 10 receives the “specific keyword” and performs an article search from the articles acquired in S102 (S105). Further, the social importance adding means 102 adds the social importance to the article extracted in S103 (S106).

  Next, the social importance calculation means 103 calculates the social importance of each article acquired in S102 (S107). For the articles searched and extracted in S105, the social importance calculated in S106 is added, and the social importance is calculated in the same manner as other articles.

  Next, the personal interest level calculation means 104 calculates the personal interest level for each article acquired in S102 according to the interest level for each genre accumulated in the client 10 (S108).

  Next, the list display article selection means 105 selects articles randomly according to the social interest level and the personal interest level calculated in S107 and S108 (S109). Although the specific algorithm of selection is not limited, it is preferable that the higher the social interest level, the easier the selection, and the higher the personal interest level, the easier the selection. The selected item is displayed on the display output means 107 (S109).

  FIG. 5 shows an example of display output by the information processing in S109. As illustrated in FIG. 5, the display output unit 107 displays a plurality of articles (18 articles in this example) in the form of a list among the acquired articles. In this example, the headlines of articles are arranged in tiles, and a screen that can be selected by clicking or the like is configured. Of these, articles that have been added by the social importance adding means 102 are indicated by “article a” and “article b” in FIG. The remaining articles 1 to 16 are randomly extracted based on social importance and personal interest without being added.

Refer to FIG. 4 again.
Next, the advertisement article list display count report means 106 counts the number of articles displayed by the display output means 107 in S109 (the number of times displayed) among the articles extracted by the search in S105, and evaluates the total result. It transmits to the server 11 (S111). The timing of S111 may be after S101 to S110 are repeated a plurality of times.

  In the evaluation server 11, the advertisement article list display count counting unit 108 totals the advertisement article list display count transmitted from each of the plurality of clients 10 (S112). Thereby, the effect of inputting the “specific keyword” to each client 10 can be measured.

  In the present embodiment, in addition to the above-described configuration, if the client 10 controls whether or not the display probability of an article searched and extracted using a specific keyword is increased, the evaluation server 11 This is preferable because it can effectively measure the advertising effect.

  In the case of this configuration, the advertisement article list display count reporting means 106 also retains control information regarding whether or not the selection probability has been increased using the specific keyword, and when the report is sent to the evaluation server 11 Control information is also sent. If it does in this way, it can distinguish and total whether the displayed article was displayed by advertisement display control, or was originally displayed because user preference and social importance were high.

  Similarly, it is possible to set a plurality of advertising plans such as “200% increase in advertisement display for 200 people” and “20% increase in advertisement display for 100 people”, and total the relationship with the actual number of display times. it can. This is useful because there are cases where a plurality of advertisers place advertisements under similar conditions and cases where they place advertisements alone.

  In addition, if the specific condition set by the advertiser is sent to the evaluation server 11, cross tabulation with the user's preference can be performed. For example, if you advertised “Laundry Detergent NC” to both users who like sports articles and users who do n’t like sports articles, it also counts which user segment pixel matches, including user preferences and interests. can do.

<Third Embodiment>
Next, a third embodiment will be described with reference to the drawings.

  In the second embodiment, as shown in FIG. 5, articles (articles a and b) that have received the addition of social importance in the list display are not added (articles 1 to 16). It is displayed as a whole. Even in the selection processing by the list display article selection means 105, the selection of articles that have been added and the selection of articles that have not been added are not separated and processed step by step, and all articles are selected in the same row. It is processed.

  On the other hand, in the present embodiment, articles that receive addition and articles that do not receive are selected separately. Articles that are subject to addition are preferably displayed on a display output using a display frame (advertisement display frame) that is different from the display frame of ordinary articles that are not subject to addition (see FIG. 8). Reference later). In the following, the “article subject to addition” will be referred to as “advertising article” by representing the above “advertising article”.

FIG. 6 shows a system configuration example of this embodiment.
As shown in FIG. 6, the system of the present embodiment includes an advertisement distribution server 13 in addition to the system configuration of the second embodiment, for example. The advertisement distribution server 13 distributes advertisements to be displayed in an advertisement frame that is displayed separately from normal articles distributed from the article distribution server 12 in the list display in the client 10.

FIG. 7 shows a functional configuration of the present embodiment.
As illustrated in FIG. 7, the client 10 includes an advertisement article search unit 101, an advertisement article selection unit 109, and an advertisement article list display count report unit 107. Further, the evaluation server 11 includes an advertisement article list display count counting unit 108. The configuration of the advertisement article search means 101, the advertisement article list display frequency report means 107, and the advertisement article list display frequency count means 108, excluding the advertisement article selection means 109, is the same as that of the second embodiment.

  Further, the client 10 includes a social importance level calculation unit 103, a personal interest level calculation unit 104, a list display article selection unit 105, and a display output unit 107 that have the same functions as those shown in FIG. Since these are not directly related to the features of this embodiment, they are not shown.

  In the present embodiment, an area for displaying an advertisement article is secured as an advertisement frame in a specific area on the list. FIG. 8 shows an example of display output. As shown in FIG. 8, in the present embodiment, articles (articles 1 to 15) that are not subject to the addition of social importance are displayed in a tiled manner in a list, while an advertisement article or an advertisement (advertisement a, advertisement b) is displayed. ) Is displayed in a different area (advertisement display frame) from articles 1-15. The size of the advertisement display frame and the number of advertisement articles displayed in the advertisement display frame are arbitrary.

  An advertisement article (that is, an article extracted by the advertisement article search means 101) is not necessarily selected as the advertisement space. Various types of advertisements are distributed to the advertisement article selection means 109 from an external advertisement server (which may be the same as the article distribution server 12). The advertisement article selection unit 109 selects the advertisement based on the user's personal interest level from among the advertisement group including the distributed advertisement and the article (advertisement article) extracted by the advertisement article search unit 101, as shown in FIG. Select an advertisement to be displayed in the advertising space shown in.

  More specifically, the advertisement article selection unit 109 selects an advertisement to be displayed in the advertisement space based on a probability calculated from an advertisement fee presented by a plurality of advertisers when selecting an advertisement. Here, in the present embodiment, the advertisement is further selected based on not only the advertisement fee but also the probability calculated from the compatibility with the specific keyword input from the evaluation server 11 to the advertisement article search means 101. Preferably, the probability used for the selection is also linked to the suitability with the user's interest based on the calculation result of the personal interest degree calculation unit 104.

  As with the second embodiment, the number of times the articles extracted by the advertisement article search means 101 whose probability of being selected has been increased by the advertisement article selection means 109 is displayed as a list is the same as that of the second embodiment. It transmits to the evaluation server 11. The advertisement article list display frequency coefficient means 108 of the evaluation server 11 totals the advertisement article list display times transmitted from the plurality of clients 10. Thereby, the effect of inputting the “specific keyword” to each client 10 can be measured.

<Fourth Embodiment>
Hereinafter, the fourth embodiment will be described with reference to the drawings.

  FIG. 9 shows a functional configuration of the system of the present embodiment. This embodiment is similar to the second embodiment, but in comparison with this, the evaluation server 11 has the advertisement article search means 101. The configuration in which the plurality of clients 10 each have the advertisement article search unit 101 as in the present embodiment is not essential, and a search using a specific keyword can be executed on the server side.

  Moreover, even if it comprises so that the advertisement article search means 101 of 3rd Embodiment may have in the evaluation server 11, the effect similar to 3rd Embodiment is brought about.

<Summary>
Hereinafter, the operational effects of the above embodiment and modified examples will be described.
According to the above embodiment, there are advantages as described below for both the advertiser on the evaluation server 11 side and the user of the client 10.

  In recent years, advertisements called native advertisements, which are different from article advertisements that are directly advertised to users by companies that are advertisers, have also become widespread. This is to have the delivery media trusted by readers comment on products and services. If the reviews are favorable, the appeal to reader users is strong. According to the above embodiment, the exposure of such a native advertisement can be increased. In addition, the effect can be obtained in a form that is in line with social importance such as the user's interests and preferences, and the number of references on the Internet and SNS.

That is, the advertiser can expose the media and the user's positive product introduction article as an advertisement without incurring the cost of preparing the advertisement creative.
The user can obtain an article that meets the interest instead of the advertisement that is not interested. In particular, in the second embodiment, since the social attention level is added and weighted, there is a high possibility that articles that do not match the user's personal interest level will not be displayed.
In particular, in the third embodiment, advertisement articles are forcibly picked up, but among a number of articles, articles that match a dictionary representing the user's interest are picked up.

  In general, an advertiser who has a financial margin can place an advertisement on a keyword that is frequently searched and increase the display probability even if the advertiser is irrelevant to his / her product. However, this method only pays advertising fees for articles unrelated to the company, and has little merit. Moreover, it is highly likely that the advertisement is useless to the user, and it is advantageous to reduce such an advertisement.

  Advertisements that are significantly limited in number compared to articles will be displayed repeatedly and are annoying for users, but it is useful to give users multi-faceted information by using articles that introduce their products positively as advertisements It is possible to behave as simple information and to promote purchase consideration, and to reduce the annoyance compared to repeatedly displaying the same advertisement.

DESCRIPTION OF SYMBOLS 10 Client 11 Evaluation server 12 Article distribution server 13 Advertisement distribution server 14 Advertisement information distribution server 101 Advertising article search means 102 Social importance addition means 103 Social importance calculation means 104 Individual interest degree calculation means 105 List display article selection means 106 Advertising article list display frequency reporting means 107 Display output means 108 Advertising article list display frequency counting means 109 Advertising article selection means

Claims (5)

A system including an advertisement distribution server for distributing advertisements and an article distribution server for distributing articles,
Importance adding means for adding importance to an article distributed from the advertisement delivery server and extracted by the article selection condition that is advertisement information;
Importance calculating means for calculating the importance of each of the articles distributed from the article distribution server ;
Selecting means for selecting one or more for display in a predetermined advertising space from among advertisements distributed from the advertisement distribution server and articles distributed from the article distribution server ;
Displaying one or a plurality of advertisements distributed from the advertisement distribution server and articles with the importance added by the importance adding means on a predetermined advertisement space, distributed from the article distribution server, Display output means for displaying an article in which the importance is not added by the importance adding means in a frame different from the predetermined advertising space;
With
The system is characterized in that the selection means selects the articles so that the articles with higher importance are more easily selected. An advertisement article search means for searching and extracting an article group distributed from the article distribution server as an article selection condition that matches a specific keyword specified by an advertiser,
The system according to claim 1, wherein the importance calculation unit increases the importance of an article extracted by the advertisement article search unit.   The system according to claim 2, wherein the advertisement article search means specifies and searches whether the article is a positive article or a negative article for a specific keyword designated by an advertiser.   The system according to claim 2, further comprising: a display frequency reporting unit that reports to an external evaluation server the number of times the article whose importance is increased by the importance calculation unit is selected by the selection unit. Display output means for displaying and outputting the article selected by the selection means;
When the display output means displays a list of articles, a display frequency reporting means for reporting to the external evaluation server the number of times the importance calculation means includes an article with an increased importance in the list display;
The system of claim 2, comprising:

Images