JP6580484B2 - Information provision management method and information provision management apparatus - Google Patents

Description

  The present invention relates to an information provision management method and an information provision management apparatus.

  In recent years, services that provide information via the Internet have become widespread as social infrastructure. Services that provide information using the Internet include blogs, Internet bulletin boards including so-called summary sites, and social network services (SNS). SNS is a member service for building a social network on the Internet. In SNS, a post for a member to transmit information is disclosed to a specific other member, and the other member views the post, adds a comment to the post, or reposts the post. And information spreads further. Conventionally, advertisement distribution using SNS has been studied (for example, see Patent Document 1).

  In Patent Literature 1, an SNS site control unit automatically creates a member user's comment by utilizing the comment function of the SNS, so that the member user does not need to fill in a promotional phrase and the like, and an advertisement with customer attraction is provided. An advertisement providing system that can be provided has been proposed.

  In the advertisement providing system of Patent Literature 1, when a member user of SNS executes a commercial transaction process corresponding to a reward target process on a commercial website provided by an advertiser server, a user registered as a related user of the user is registered. A comment including link information to the advertiser server is created and displayed on the My Page without manual posting by the user.

JP 2012-242976 A

  However, in the advertisement providing system of Patent Document 1, since automatic comment posting to the SNS server is performed in response to processing at the advertiser server, information posted by the member to the SNS, in particular introduction of food and purchased products There is a problem that it is not possible to automatically post an appropriate advertisement according to the image used in the above.

  The present invention has been made in view of such problems, and an object thereof is to provide an information provision management method and an information provision management apparatus capable of effectively providing information suitable for an image included in a post.

  The information provision management method of the present invention is executed by a computer device including a network control unit, a storage unit, and a central control unit connected to a network, and the central control unit is configured to use a plurality of posted images included in a plurality of posts. A step of extracting, as a target image, a post image including a related image related to a registered image stored in advance in the storage unit and not including the registered image, from the image group A configured; and the central control unit Comprises associating and storing the target image and post-related information related to the post including the target image as target information in the storage unit.

  With this configuration, a post image including a related image related to a registered image and not including a registered image is extracted as a target image from the first image group A composed of a plurality of post images included in a plurality of posts. Since the target information is obtained by associating the target image and the post related information related to the post including the target image, the information related to the registered image can be effectively transmitted based on the target information. .

  The information provision management method of the present invention further includes a step of determining the related image from a plurality of candidate images, wherein the number of posted images including each of the plurality of candidate images, and the central control unit includes a plurality of images on the network. Based on the image group A ′ composed of a plurality of posted images included in the post, the number of posts including a word corresponding to the registered image, the number of posts including a word indicating emotion, and included in the post Based on the affinity, which is a parameter indicating the relationship between the candidate image and the registered image, calculated based on at least one of the appearance probabilities of other words that are related to a specific word It is preferable to determine the related image from the candidate images.

  In the information provision management method of the present invention, it is preferable that the central control unit further includes a step of controlling the network control unit to provide information about the target image via the network. .

  In the information provision management method of the present invention, it is preferable that the central control unit further includes a step of setting a price of information provision to the target image based on the target information.

  In the information provision management method of the present invention, it is preferable that the price is set based on an influence degree of a contributor of the posting.

  In the information provision management method of the present invention, it is preferable that the central control unit further includes a step of calculating a degree of compatibility of the posting poster with respect to the registered image.

  In the information provision management method of the present invention, it is preferable that the posts constituting the image group A are collected before posting.

  An information provision management device according to the present invention includes a network control unit, a storage unit, and a central control unit connected to a network, and the central control unit includes an image group composed of a plurality of posted images included in a plurality of posts. From A, the post image including the related image related to the registered image stored in advance in the storage unit and not including the registered image is extracted as a target image, and the post including the target image and the target image The post-related information related to is stored as target information in the storage unit.

  With this configuration, a post image including a related image related to a registered image and not including a registered image is extracted as a target image from the image group A including a plurality of post images included in a plurality of posts. Since the target information associated with the post-related information related to the post including the image is obtained, the information related to the registered image can be effectively provided based on the target information.

  According to the present invention, information related to a registered image can be effectively provided based on target information, so that information suitable for an image included in a post can be effectively provided.

It is explanatory drawing which shows the outline of the information provision management system which concerns on this Embodiment. It is a block diagram which shows the server in the information provision management system which concerns on this Embodiment. It is a block diagram which shows the terminal in the information provision management system which concerns on this Embodiment. It is a functional block diagram which shows the function of the advertisement server which concerns on this Embodiment. It is a flowchart figure which shows each process of the registration process of the advertiser in the information provision management method concerning this Embodiment. It is explanatory drawing which shows an example of advertiser DB in the information provision management method concerning this Embodiment. It is a schematic diagram which shows the specific example of the registration image in the information provision management method which concerns on this Embodiment. It is a sequence diagram which shows each process of the registration process of the permission member in the information provision management method concerning this Embodiment. It is explanatory drawing which shows an example of the permission member DB in the information provision management method which concerns on this Embodiment. It is a flowchart figure which shows each process of the collection process of the 4th image group A 'in the information provision management method concerning this Embodiment. It is explanatory drawing which shows an example of contribution image DB in the information provision management method concerning this Embodiment. It is a schematic diagram which shows the specific example of the contribution image which belongs to the 5th image group B 'in the information provision management method concerning this Embodiment. It is a flowchart figure which shows each process of the extraction process of the 5th image group B 'in the information provision management method concerning this Embodiment. It is explanatory drawing which shows an example of candidate image DB in the information provision management method which concerns on this Embodiment. It is a flowchart figure which shows each process of the extraction process of 6th image group C1, C2 ... Ck in the information provision management method concerning this Embodiment. It is a schematic diagram which shows the specific example of the posting image which belongs to the 6th image group C1, C2 ... Ck in the information provision management method concerning this Embodiment. It is a flowchart figure which shows each process of the determination process of the affinity image in the information provision management method concerning this Embodiment. It is a flowchart figure which shows each process of the collection process of the 1st image group A in the information provision management method concerning this Embodiment. It is a flowchart figure which shows each process of the extraction process of the 2nd image group B in the information provision management method which concerns on this Embodiment. It is a flowchart figure which shows each process of the extraction process of the 3rd image group D in the information provision management method concerning this Embodiment. It is a flowchart figure which shows each process of the creation process of the affinity opportunity list | wrist in the information provision management method which concerns on this Embodiment. It is explanatory drawing which shows an example of advertisement object DB (affinity opportunity list) in the information provision management method which concerns on this Embodiment. It is explanatory drawing which shows an example of stock DB in the information provision management method which concerns on this Embodiment. It is a schematic diagram which shows the specific example of the target image in the information provision management method which concerns on this Embodiment. It is a flowchart figure which shows each process of the setting process of the minimum bid amount of an advertisement in the information provision management method concerning this Embodiment. It is a sequence diagram which shows each process of the bid process of the advertisement in the information provision management method which concerns on this Embodiment. It is a flowchart figure which shows each process of a calculation process of the compatibility with the registration image of the contributor in the information provision management method concerning this Embodiment. It is a sequence diagram which shows each process of the delivery process of the advertisement in the information provision management method which concerns on this Embodiment. It is a flowchart figure which shows each process of the transmission process of the posting with an advertisement in the information provision management method concerning this Embodiment. It is a schematic diagram which shows the specific example of the advertisement in the information provision management method which concerns on this Embodiment.

  Hereinafter, an information provision management method and an information provision management device according to embodiments of the present invention will be described in detail with reference to the accompanying drawings. The information provision management method and the information provision management device according to the present invention are not limited to the following embodiments, and can be appropriately modified and implemented within the scope of the present invention.

  “Information” in the information provision management method of the present invention includes, for example, advertisements, contents, and PR. In the present embodiment, a case where the present invention is applied to advertisement provision management will be described as an example. FIG. 1 is an explanatory diagram showing an outline of the information provision management system according to the present embodiment. As shown in FIG. 1, in the information provision management system 1, an advertisement server 2, an SNS server 3, a settlement server 4, a member terminal 5, and an advertiser terminal 6 that constitute an example of an information provision management device are: They are connected to each other via a network 7 so that they can communicate with each other.

  The network 7 is, for example, the Internet, and includes an Internet backbone network, a LAN, a mobile communication network, Wi-Fi, and the like.

  Here, for simplicity of explanation, one member terminal 5 and advertiser terminal 6 are illustrated, but there are actually many.

<Server>
As described above, the information provision management system 1 according to the present embodiment includes the advertisement server 2, the SNS server 3, and the settlement server 4. FIG. 2 is a block diagram showing a server in the information provision management system according to the present embodiment.

  As shown in FIG. 2, the server 100 as an example of the computer apparatus of the present invention includes a central control unit 101, a network control unit 102, a storage device 103, a memory 104, and a power supply unit 105.

  The central control unit 101 controls each unit of the server 100. Specifically, the central control unit 101 includes a CPU (Central Processing Unit) and the like, and performs various control operations using the memory 104 according to the OS and various processing programs.

  The network control unit 102 performs data communication and the like via the network 7.

  The storage device 103, which is an example of the storage unit of the present invention and is an external storage device, is configured by, for example, a nonvolatile memory and stores various programs and data necessary for the operation of the central control unit 101.

  The memory 104, which is an example of the storage unit of the present invention, is composed of, for example, a DRAM (Dynamic Random Access Memory) and the like, and includes a buffer memory that temporarily stores data processed by the central control unit 101 and the like, and a central control It functions as a working memory for the unit 101 and the like.

  The power supply unit 105 is connected to an external power source in order to supply necessary power to the server 100.

  In the server 100 described above, the advertisement server 2, the SNS server 3, and the settlement server 4 can be realized by appropriately preparing programs and data according to the purpose.

<Terminal>
FIG. 3 is a block diagram showing a terminal in the information provision management system according to the present embodiment. As shown in FIG. 3, the terminal 200 includes a central control unit 201, a network control unit 202, a display unit 203, a UI device 204, a GPS reception unit 205, a camera unit 206, a storage device 207, a memory 208, and a power supply unit 209. I have.

  The central control unit 201 controls each unit of the terminal 200. Specifically, the central control unit 201 includes a CPU and the like, and uses the memory 208 to perform various control operations according to an operating system (OS) and various processing programs.

  The network control unit 202 performs data communication and the like via the network 7.

  The display unit 203 displays various information generated by the central control unit 201.

  A UI (user interface) device 204 receives user operations, and includes buttons, keys, a touch panel, and the like. The UI device 204 outputs a predetermined operation signal to the central control unit 201 according to these operations.

  The GPS receiving unit 205 receives a GPS signal from an artificial satellite and outputs a geographical position (for example, coordinates) to the central control unit 201.

  The camera unit 206 is an example of an image input unit, and includes a lens and an optical element for capturing a still image or a moving image.

  The storage device 207 is configured by, for example, a nonvolatile memory (flash memory) or the like, and stores various programs and data necessary for the operation of the central control unit 201.

  The memory 208 is constituted by, for example, a DRAM or the like, and functions as a buffer memory for temporarily storing data processed by the central control unit 201 or the like, or a working memory for the central control unit 201 or the like.

  The power supply unit 209 is connected to a battery or an external power source in order to supply necessary power to the terminal 200.

  In the terminal 200 described above, various functions described below can be realized by appropriately preparing application programs and data according to the purpose.

  Hereafter, each server and terminal which comprise the information provision management system 1 which concerns on this Embodiment are demonstrated.

<Advertising server>
FIG. 4 is a functional block diagram showing functions of the advertisement server according to the present embodiment. The advertisement server 2 is an example of the information provision management device of the present invention, and includes an advertisement target management unit 21, a placement management unit 22, and an advertisement provision unit 23.

  The advertisement target management unit 21 includes a related image determination unit 24, an AO list generation unit 25, and an inventory management unit 26.

  The related image determination unit 24 relates to a registered image (described later) received from the advertiser terminal 6 from the fourth image group A ′ composed of images included in posts on the network 7 (hereinafter referred to as posted images). The process which determines the related image to perform is performed.

  The AO list generation unit 25 extracts, as a target image, a post image that includes a related image and does not include a registered image from the first image group A that includes a plurality of post images included in a plurality of posts. And an affinity opportunity (hereinafter referred to as AO) list in which advertising target information is registered, which is an example of target information related to an information provision target, in which the target image and post-related information related to a post including the target image are associated. And processing to execute.

  The inventory management unit 26 performs processing for extracting and managing posts including target images by authorized members (described later) from target images included in the AO list.

  Further, the placement management unit 22 first sets a price for the inventory. In addition, the manuscript management unit 22 controls the network control unit 102 to transmit a manuscript request including a price via the network 7, and from the advertiser, which is an aspect of the information provision request of the present invention. And a process of receiving a placement request via the network 7.

  Further, the advertisement providing unit 23 executes a process of transmitting an advertisement targeted for the inventory on the network 7 based on the submission request received by the submission management unit 22. Further, the advertisement providing unit 23 bills the settlement server 4 for an advertisement reward or an affiliate's consideration for the member described later based on the feedback (feedback) (described later) from the member to the advertisement distribution or posting by the SNS server 3. Request payment.

<SNS server>
The SNS server 3 has a function for providing a social network service (SNS). More specifically, the central control unit 101 (see FIG. 2) of the SNS server 3 performs all or part of the following processing.
・ Process to receive and publish a post from the member terminal 5 operated by the member (hereinafter referred to as “main post”) ・ Process to receive a comment on the main post from the member terminal 5 operated by another member and publish it ・ Other members The member terminal 5 operated by the other member receives a request for re-posting (for example, called retweet or share) from the member terminal 5 operated by the member and republishing the main posting as a publisher. Processing to receive responses (eg “Favorites”, “Like”, “Like”) to main posts from

  The above-described processing is an example of processing performed by the SNS server 3, and is not particularly limited.

The SNS server 3 has a function of managing member related information. Member-related information is, for example, all or part of the following.
・ Member ID
・ Password, member name, member attribute information (gender, age, occupation, etc.)
・ Number of posts, number of public addresses, and member IDs of public addresses
・ Number of publishers, member ID of publisher

  Note that the publication destination is another member to whom the posting is published, for example, a Twitter (registered trademark) follower, Facebook (registered trademark), or a LINE (registered trademark) friend. In addition, the publication source is another member whose member is a disclosure destination. For example, in Twitter, the member is another member that the member is following.

  The specific implementation of the SNS server 3 can be realized by a known technique and can be easily understood by those skilled in the art.

<Payment server>
The settlement server 4 has a function of billing the advertiser for advertisement rewards and paying the compensation of the affiliate to the member based on the settlement request from the advertisement server 2. The specific implementation of the payment server 4 can be realized by a known technique and can be easily understood by those skilled in the art.

<Member terminal>
The member terminal 5 is a terminal 200 operated by a SNS member.

  In the member terminal 5, an application program for SNS (hereinafter referred to as an SNS application) is executed by the central control unit 201, and various processes such as posting, commenting, reposting, and echoing to the SNS server 3 can be executed. It is configured as follows.

  Hereinafter, processing performed by the SNS application executed on the terminal 200 will be described. First, the SNS application executes a process for creating a post in response to a member operation performed via the UI device 204. Posts to be created include, for example, text input from the UI device 204, geographical position (post position information) input from the GPS receiving unit, and image (posted image) input from the camera unit 206. included.

  In addition, the SNS application controls the network control unit 202 to execute a process of transmitting a post to the SNS server 3 via the network 7.

  In addition, the SNS application executes processing for creating and transmitting a comment, requesting re-posting, and notifying of an echo in response to a member operation performed via the UI device 204.

  Further, the SNS application receives a post published by the SNS server 3, displays the post on the display unit 203, and executes processing.

  Here, the SNS application has been described assuming a native application program that is installed in advance in the storage device 207 of the terminal 200 and executed. However, a function equivalent to the SNS application may be implemented using a Web application. That is, in the terminal 200, the central control unit 201 executes a browser program. The browser program interprets and executes a Web application program created by combining, for example, HTML5, Javascript (registered trademark), CSS, and the like. As a result, the browser program requests the server-side program executed on the SNS server 3 for various processes, receives the results, and creates and transmits the post as described above.

  Various processes of the SNS application may be realized by a hybrid application that combines the native application and the web application as described above.

  Further, as will be described later, the member terminal 5 is configured to execute processing for permission of advertisement distribution and authentication confirmation. Furthermore, as will be described later, the member terminal 5 is configured to execute a process of depositing a post before publication in the AO list generation unit 25 of the advertisement server 2.

  That is, in the member terminal 5, the browser program executed by the central control unit 201 interprets and executes the Web application program provided by the advertisement server 2 and the SNS server 3. As a result, the browser program requests the server-side program executed on the advertisement server 2 and the SNS server 3 to perform processing for receiving a post before publication, permission for advertisement distribution, and authentication confirmation.

  Here, the member terminal 5 may execute processing for transmission of a post before publication, permission for advertisement distribution, and authentication confirmation by a dedicated native application or a hybrid application instead of a browser program. In addition, the SNS application may execute processing for sending a post before publication, permission for advertisement distribution, and authentication confirmation.

  In the present embodiment, the member terminal 5 is assumed to be a smartphone or a tablet, but is not particularly limited, and may be various portable terminals such as a smart watch or a smart glass having a communication function. Needless to say, it is good. However, the terminal 200 is not limited to a portable terminal, and includes, for example, all devices connected to the network 7 such as a personal computer.

<Advertiser terminal>
An advertiser is an example of a provider of information of the present invention. The advertiser terminal 6 is a terminal 200 operated by the advertiser. The advertiser terminal 6 communicates with the advertisement server 2 and the settlement server 4 via the network 7 and, as will be described later, registration of advertiser information and images, display of advertisement placement requests, advertisement placement requests, It is configured to execute processing such as payment of rewards.

  In the advertiser terminal 6, the browser program executed by the central control unit 201 interprets and executes the Web application program provided by the advertisement server 2 and the settlement server 4. As a result, the browser program requests the server-side program executed on the advertisement server 2 and the settlement server 4 to execute the various processes described above.

  Here, the advertiser terminal 6 may execute the various processes described above by a dedicated native application or a hybrid application instead of the browser program.

  In the present embodiment, the advertiser terminal 6 is assumed to be a personal computer, but it goes without saying that various portable terminals such as smartphones and tablets may be used.

<Process flow>
Hereinafter, the information provision management method performed by the information provision management system 1 according to the present embodiment will be described in detail.

  The information provision management method according to the present embodiment includes (A) generation of advertisement target information, (B) determination of related images, (C) management of inventory, (D) management of advertisement placement, and (E) advertisement. It can be broadly divided into distribution management.

(A) Generation of Advertisement Target Information In the information provision management method according to the present embodiment, in the advertisement server 2, the central control unit 101 of the server 100 (see FIG. 2) uses a plurality of posted images included in a plurality of posts. A step of extracting a post image including a related image related to a registered image stored in advance in the storage device 103 or the memory 104 and not including a registered image as a target image from the configured first image group A; Storing the target image and the post-related information related to the post including the target image and storing the target image as advertisement target information in the storage device 103 or the memory 104.

  Here, posting means a unit of information that is uploaded by a member who is a user to an information providing service using the Internet and is provided to other members.

  Posting includes, for example, posting to an SNS, posting to a blog (also called a diary or article), and posting to an Internet bulletin board (also called writing).

  In the present embodiment, the posting is an SNS posting, and includes Twitter Twitter and Twitter, Facebook submissions and comments, and the like.

  SNS is an aspect of communication between social networks or individuals, and the present invention is not limited to SNS, and can be suitably applied to information sharing services between communities or between individuals.

  Further, the posted image is an image included in the posted. The image includes a still image and a moving image.

  The registered image is, for example, an image provided by an advertiser and is an image related to the advertisement. For example, the image of the product itself to be advertised, the name of the product and service used in the advertisement, Includes images of company names such as logos and characters. As will be described later, in the present embodiment, it is assumed that the registered image is transmitted to the advertisement server 2 by the advertiser using the advertiser terminal 6, but a person other than the advertiser (advertising agent) Etc.) may be transmitted.

  Post-related information is information related to a post, such as a post ID, post text information, post date, post location, number of views, number of viewers, and a member who made a post (hereinafter referred to as a poster). Member-related information). Here, post text information refers to a character string included in or attached to a post such as a post body, a title, and a tag.

  The related image refers to an image that has some relationship with the registered image. For example, a registered image co-occurrence image is included. Here, the co-occurrence image refers to an image that is expected to be associated from a predetermined target (image). For example, an image that is highly relevant to a predetermined object, an image that may be used at the same time, or the like.

  Co-occurrence images include, for example, images of other products associated with registered images, images of associated products and services, logos of company names, images of characters, etc., scenery associated with registered images, etc. Including images.

  The target image includes a related image extracted from the first image group A composed of a plurality of posted images included in a plurality of posts and related to the registered image, and does not include the registered image. It is an image. Members who see related images included in the target image are expected to be associated with the registered image, so by adding advertisements to such posted images or posts containing posted images, excellent advertising effects are expected Is done.

  The posted image constituting the first image group A may be included in a post before being released by the SNS server 3 or may be a posted image that has already been released.

  If the advertisement server 2 collects posts from SNS members in advance, configures the first image group A with a plurality of posted images included in the plurality of posts, and extracts a target image from this, the post before release After an advertisement is added to an image or a post, the post can be transmitted to the SNS server 3 and made public.

  In addition, if the posted images included in the posted posts are collected from the SNS server 3 to form the first image group A, and the target images are extracted from this, comments including advertisements are posted on the posted posts. You can post or repost by adding an advertisement to the post.

  The advertisement target information is one aspect of the “target information” of the present invention, and associates the target image and the posting related information of the post including the target image. As described later, based on the advertising target information, it is possible to provide information for determining the price of the advertisement and determining the placement by the advertiser, and easily distribute the advertisement and settle the advertising reward based on the advertising target information. it can.

  In the information provision management method according to the present embodiment, in the advertisement server 2, the central control unit 101 of the server 100 (see FIG. 2) sends the posted image including the registered image from the first image group A to the second image. A step of extracting as an image group B, and a step in which the central control unit 101 extracts a post image belonging to the first image group A including a related image as a third image group D; The step of extracting the target image from one image group A is preferably performed by excluding the posted image belonging to the third image group D and belonging to the second image group B from the third image group D.

  The second image group B is composed of posted images including at least a registered image. That is, the second image group B includes a posted image (a) including only a registered image, a posted image (b) including a registered image and a related image, and a post including a subject image and a registered image other than the related image. Image (c) is included.

  The third image group D is composed of posted images including at least related images. That is, the third image group D includes a posted image (d) including only related images, a posted image (b) including registered images and related images, and a posted image including images of subjects other than registered images and related images. (E) is included.

  Therefore, the posted image belonging to the third image group D and belonging to the second image group B, that is, the posted image (b) including the registered image and the related image is excluded from the third image group D (d + b + e). Thus, the target image (d + e) can be extracted from the first image group A.

  Although the target image can be extracted from the first image group A as described above, it is not necessarily limited to the processing method described here.

(B) Determination of related image In the information provision management method according to the present embodiment, the advertisement server 2 further includes a step in which the central control unit 101 determines a related image from a plurality of candidate images, and the fourth image Based on the group A ′, the number of posted images including each of the plurality of candidate images, the number of posts including a word corresponding to the registered image, the number of posts including a word indicating emotion, and a specific included in the post Based on the affinity, which is a parameter indicating the relationship between the candidate image and the registered image, calculated based on at least one of the appearance probabilities of other words that are related to the word. It is preferable to carry out the determination of the relevant image.

  In the present embodiment, the degree of affinity (affinity, affinity, affinity) is a parameter indicating the relationship between a candidate image and a registered image.

  The candidate image is an image that is considered to be related to the registered image, but may be an image that is expected to be related to the registered image even though it is not actually related to the registered image. Without specifying the registered image, for example, a product itself, a product and service name, a company name logo, a person, a character, a landmark, and a landscape image can be prepared and used as candidate images.

  In the present embodiment, the degree of affinity can be calculated based on one or more of the following characteristics. For example, the number α of posts having one or more of the following characteristics is used. Further, the appearance probability of posts having the following characteristics may be used. The appearance probability may be calculated by Bayesian statistics.

-Including each of the plurality of candidate images The posted image including the candidate image may be a posted image including both the registered image and the candidate image. That is, the number of posted images including candidate images may be the number (n) of posted images including both registered images and candidate images.

-Post including the word corresponding to the registered image The word (word) corresponding to the registered image is, for example, the product name of the product included in the registered image or the service name of the service related to the registered image, and is also called a brand word. .

-Posts that include words that indicate feelings Words that indicate feelings are words that indicate sentiment, such as emotion, emotion, sentiment, etc. Negative words that express feelings such as hate, sadness, anger, etc. that express a feeling of friendliness or a favorable feeling for the product name or product related to the registered image, such as “delicious” Is a word.

・ Specific words included in a post and other words that are related to a specific word Dependently analyze a sentence included in the posted text information of a post, identify one word in the sentence, and relate to that word Other words in relation can be counted, and the appearance probability of other words in relation to a specific word can be calculated. A known technique can be used for dependency analysis.

  The related image determination unit 24 shown in FIG. 4 controls the network control unit 102 (see FIG. 2), and from the SNS server 3 (see FIG. 1) via the network 7 (see FIG. 1) during a predetermined period. Post images included in a plurality of corresponding posts are collected to obtain a fourth image group A ′. Next, the related image determination unit 24 is a parameter indicating the relevance between the candidate image and the registered image using at least one of the above-described number α based on the fourth image group A ′. Calculate the affinity. When the posted text information is used, such as the number of posts including the word corresponding to the registered image or the number of posts including the word indicating the emotion among the number α, the related image determination unit 24 selects the posted image as When collecting or separately, post text information of a plurality of posts is acquired from the SNS server 3 and used for the calculation of the number α described above.

  Next, based on the calculated affinity, the related image determination unit 24 selects, for example, those having an affinity greater than or equal to a predetermined value or less than a predetermined value from among a plurality of candidate images. Are determined as related images.

  The affinity is calculated using the number α as described above. Although the calculation method is not particularly limited, for example, any one or a combination of the following methods is used.

(I) When the affinity is calculated based on the total number (N) of posted images including registered images and the number (n) of posts corresponding to posts including registered images and candidate images. Will be described in detail later.

(Ii) When obtaining the affinity by taking at least two correlations among the above-mentioned number α For example, for a certain candidate image, a fourth composed of a plurality of posted images included in a plurality of posts to the SNS server 3 The number (n1) of posted images including registered images and candidate images included in the image group A ′, and the number (n1 ′) of posts including a word (for example, ABC soda) corresponding to the registered image among the plurality of posts described above. ), For example, can be calculated for a plurality of periods and obtained as the affinity.

(Iii) When the above coefficients are weighted by clustering analysis and the affinity is obtained Clustering analysis is one of data analysis methods, and a cluster of data is divided into subsets (clusters), and each cluster is analyzed. Make certain the data contained in each subset have some common features. Common features include, for example, the gender or age of the poster, and use the age group and gender classification used for marketing (for example, M1, M2, M3, F1, F2, F3). be able to.

  For example, in the fourth image group A ′, among posts including a candidate image, the number of posts by the F1 layer poster (n1 ′), the number of posts by the F2 layer poster (n1 ″), and F3 The number of posts (n1 ′ ″) by the posters of the layer can be obtained, different weightings can be performed, and the affinity can be obtained. Thereby, it is possible to calculate the affinity degree in consideration of the age group and the sex.

(Iv) In the case where the affinity is obtained from the parameter estimation in the regression analysis for the above-mentioned coefficients, multinomial logistic regression analysis and multiple regression can be used for the regression analysis, but are not limited to these. is not. In the information provision management method according to the present embodiment, the affinity may be analyzed as expressed by the above-mentioned number α.

  Further, one aspect of the present embodiment will be described in detail. In the information provision management method according to the present embodiment, the step of determining a related image from a plurality of candidate images includes a registered image from the fourth image group A ′. A step of obtaining a fifth image group B ′ by extracting a posted image including the image, and the central control unit 101 extracts a posted image including each of a plurality of candidate images from the fifth image group B ′, and a plurality of candidates For each image, a step of storing the sixth image group C1, C2... Ck (k is a natural number of 2 or more) in the storage device 103 or the memory 104, and the central control unit 101 stores the fifth image group B ′. Based on the number N of posted images and the number n of posted images belonging to each of the sixth image groups C1, C2,... Ck, the affinity represented by n / N × 100 (%) is calculated. It is preferable to comprise .

  It is preferable to determine the related image used for extracting the target image as follows. First, candidate images are prepared. Candidate images may be input by an operator operating the advertisement server 2 as images that are likely to be related to the registered image, or the advertisement server 2 prepares a plurality of images in advance, and the operator operates from among them. May be selected (check box format, etc.).

  The candidate image does not necessarily have to be input in advance by the operator. For example, the related image determination unit 24 may control the network control unit 102 and use a plurality of images automatically collected from a plurality of Web sites (not shown) on the network 7 as candidate images.

  In addition, as a candidate image, an image registered in a brand database (a database in which product images and logos are registered) may be used.

  On the other hand, in the present embodiment, post images are collected from posts to the SNS server 3 to prepare a fourth image group A ′. Therefore, it is assumed that post images are collected from posts already published by the SNS server 3. Therefore, as described above, when the first image group A is composed of posted images included in the post before release, the first image group A and the fourth image group A ′ are: Not identical. On the other hand, when the first image group A is composed of posted images included in a post that has already been published, the first image group A and the fourth image group A ′ are the same. It does not necessarily have to be the same.

  Next, a posted image including a registered image is extracted from the fourth image group A ′ to obtain a fifth image group B ′. The fifth image group B ′ is composed of posted images including at least a registered image. That is, the fifth image group B ′ includes a posted image (a) including only the registered image, a posted image (f) including the registered image and the candidate image, and a subject other than the candidate image and an image of the registered image. The posted image (g) is included.

  Next, the central control unit 101 extracts a posted image including each of a plurality of candidate images from the fifth image group B ′ (a + f + g), and the sixth image group C1, C2,. It is stored in the storage device 103 or the memory 104 as Ck (k is a natural number of 2 or more).

  For example, the sixth image group Ck for a certain candidate image includes a posted image (g) including the candidate image and the registered image.

  Thereafter, the central control unit 101 selects at least one of the sixth image groups C1, C2,... Ck based on the number of posted images belonging to the sixth image group C1, C2,. Corresponding candidate images are determined as related images.

  It is assumed that the related image and the registered image have good compatibility. Therefore, a method for determining related images based on affinity is examined.

  The number (n) of posted images belonging to each of the sixth image groups C1, C2,... Ck represents the affinity (affinity) between the registered image and the candidate image. For example, when the number (n) of a certain candidate image (A) and another candidate image (B) is compared, it is considered that the larger the number (n), the better the compatibility with the registered image.

  The candidate images having a large number (n) are not necessarily determined as related images, but n = 1 candidate images having a small number (n) may be determined as related images. This is because an unexpected association between the candidate image and the registered image that seems to have low compatibility at first glance is found, and an unexpected advertisement opportunity may be found.

  In the present embodiment, the determination of the related image is, for example, the total number N of posted images belonging to the fourth image group A ′, and the number of posted images included in the sixth image group C1, C2,. n, based on the affinity (compatibility) calculated by n / N × 100 (%), the affinity is greater than or equal to a predetermined value, less than a predetermined value, or within a predetermined numerical range A thing is selected, and a candidate image corresponding to it is determined as a related image.

(C) Inventory management In the present embodiment, among SNS members, a target image is included in a posted image included in a post made by a member who is permitted to use the post for advertisement distribution (hereinafter referred to as a licensed member). limit. That is, the inventory management unit 26 extracts posts from authorized members from the advertisement target information and manages them as inventory. Here, the inventory refers to a target for advertising.

(D) Advertisement Placement Management In the information provision management method according to the present embodiment, the advertisement server 2 further includes a step in which the central control unit 101 sets the price of the advertisement on the target image based on the advertisement target information. It is preferable to do. Here, the price of the advertisement is an aspect of the information provision price of the present invention. The price includes, for example, a designated price, a minimum price, a maximum price, an average price, and a reference price. The price may be a specific amount or a specific amount range.

  Further, the central control unit 101 may output the price of an advertisement, which is an aspect of the information provision price. For example, the central control unit 101 of the advertisement server 2 may output an advertisement price by controlling an output device such as a display unit (not shown) or a printer unit (not shown). Further, the central control unit 101 may control the network control unit 102 to transmit via the network 7.

  Further, in this case, the central control unit 101 controls the network control unit 102 to transmit a placement request including at least advertisement target information via the network 7, and the central control unit 101 includes a plurality of A step of receiving a placement request from an advertiser via the network 7, a step of the central control unit 101 providing an advertisement targeted for the target image based on the placement request via the network 7, It is preferable to further comprise.

  Internet advertisements include reservation-type advertisements and operational advertisements. Reservation-type advertisements are advertisements such as mass advertisements that have a predetermined price, period, and content to be posted (posted surface, delivery amount, posted content, etc.). On the other hand, the operational advertisement is an advertisement in which a placement method and a bid price are not changed, but a placement method is optimized while a specific advertisement space is not fixedly purchased. Operational advertisement uses an advertising technique that automatically or immediately supports advertisement optimization by a platform using ad technology that processes a large amount of data. Typical examples of the operational advertisement include an ad exchange, an SSP (Supply-Side Platform) and a DSP (Demand-Side Platform) in addition to a search-linked advertisement.

  The present invention can be applied to both reservation-type advertisements and operation-type advertisements. As an application to reservation type advertisements, for example, using member-related information included in the advertising target information, a post image included in a specific member's post is used as an advertising space, a price is set for this, and advertising is performed. You can call for submissions from the Lord.

  On the other hand, as an application to the operational advertisement, the following processing can be performed. In the advertisement server 2, first, the central control unit 101 sets a minimum bid amount for an advertisement to the target image based on the advertisement target information. Next, the central control unit 101 controls the network control unit 102 to transmit a bid request, which is an example of a placement request, including the advertisement target information and the minimum bid amount via the network 7. Next, the central control unit 101 controls the network control unit 102 to receive bid information, which is an example of a placement request, from a plurality of advertisers via the network 7. Thereafter, the central control unit 101 determines one of a plurality of advertisers as a successful bidder based on the bid information.

  It is preferable that the above-described advertisement price (including the minimum bid amount) is set based on the degree of influence of the contributor of the posting. The degree of influence of the poster is an index indicating how much or how much the contribution made by the poster has an effect on the people. The influence level of the poster is, for example, the member attribute information (gender, age, occupation, etc.) of the poster, the number of posts, the number of publication destinations, the number of reposts, the number of reflections on the posts, the member ID of the publication destination, the publication source The number and the member ID of the publishing source can be used as they are, or can be calculated based on them.

  In addition, the setting of the price of the advertisement based on the poster is, for example, the degree of influence (for example, the number of subscribers of newspapers, magazines, etc., the number of page views (PV) of the homepage) that is the same as or similar to the degree of influence of the poster It is possible to determine an existing medium and set the advertisement price of the medium as the advertisement price for the target image.

(E) Management of advertisement distribution In the present embodiment, the advertisement of the publisher (successful bidder) is distributed on the target image corresponding to the advertisement target information based on the advertisement target information.

  The advertisement targeted for the target image includes, for example, adding an advertisement to the target image and adding an advertisement to a post including the target image.

  In addition, a post image included in a post before the post is published on the SNS server 3 (a pre-post described later) is collected, a target image is extracted based on the collected post image, and an advertisement is added to the target image. , You can add it to the post. Further, a post image may be collected from a post published on the SNS server 3, a target image may be extracted based on the collected post image, and an advertisement may be added to the target image or added to the post. In this case, the posting may be reposting (retweet) or a comment on the posting.

  Hereinafter, the information provision management method in the information provision management system 1 according to the present embodiment will be described in more detail with reference to the drawings.

<Preparation>
FIG. 5 is a flowchart showing the steps of the advertiser registration process in the information provision management method according to the present embodiment.

  In the advertisement server 2, the manuscript management unit 22 receives the advertiser information from the advertiser terminal 6 via the network 7 (S1). Similarly, the manuscript management unit 22 receives image data of a registered image from the advertiser terminal 6 (S2), and calculates a feature amount of the received image data (S3). Further, the manuscript management unit 22 registers the advertiser information, the image data, and the feature amount in the advertiser DB (S4).

  When the image is a moving image, the feature amount is calculated as follows. The moving image is decomposed into a plurality of frame per second (fps) and stored, and a feature amount is calculated for each fps.

  FIG. 6 is an explanatory diagram showing an example of the advertiser DB in the information provision management method according to the present embodiment. The advertiser DB includes, for example, advertiser information such as an advertiser ID, an advertiser name, a payment-related information such as a billing destination, advertisement creation information (creative information) necessary for preparing a post with an advertisement described later, an image Registered image-related information such as ID, image name, image data, and feature amount is registered.

  FIG. 7 is a schematic diagram showing a specific example of a registered image in the information provision management method according to the present embodiment. FIG. 7A shows a case where the registered image is a product image of a soft drink, and FIG. 7B shows a case where the registered image is a brand logo of a soft drink.

  FIG. 8 is a sequence diagram showing each step of the licensed member registration process in the information provision management method according to the present embodiment. As shown in FIG. 8, the member terminal 5 (see FIG. 1) notifies the stock management unit 26 (see FIG. 4) of the advertisement server 2 of permission for advertisement distribution (S <b> 11). The inventory management unit 26 that has received the notification transmits an authentication request to the SNS server 3 (S12). The SNS server 3 transmits to the member terminal 5 an authentication confirmation as to whether or not the member-related information held by the SNS server 3 is permitted to be disclosed or used (S13). When the member accepts the authentication at the member terminal 5 (OK) (S14), the SNS server 3 transmits the member related information to the inventory management unit 26 (S15). The inventory management unit 26 registers the received member-related information in the licensed member DB.

  FIG. 9 is an explanatory diagram showing an example of a licensed member DB in the information provision management method according to the present embodiment. As shown in FIG. 9, the licensed member DB includes, for example, a member ID, a password, a member name, the number of followers, the number of posts, a member ID of another member to be followed, an account number used for payment of the affiliate consideration, etc. Member-related information such as information is registered.

<Collection of Fourth Image Group A ′>
FIG. 10 is a flowchart showing each step of the collection process of the fourth image group A ′ in the information provision management method according to the present embodiment. As illustrated in FIG. 10, the related image determination unit 24 (see FIG. 4) of the advertisement server 2 receives a search period designation (S <b> 21). The search period is specified by, for example, an operator. Next, the related image determination unit 24 collects post images included in posts published from the SNS server 3 within a specified period (S22). Posting images can be collected using a known technique called a crawler. Further, the related image determination unit 24 collects post related information of posts from the SNS server 3 (S23). For convenience, collection of posted images and collection of post-related images are described as separate steps, but they may be performed simultaneously or the order may be changed. Further, the related image determining unit 24 associates the collected posted image and the posted image related image with each other and registers them in the posted image DB (S24).

  In the present embodiment, the related image determination unit 24 of the advertisement server 2 collects the posted images from the SNS server 3, but the subject of collecting the posted images is not limited to the advertisement server 2. That is, the method by which the operator of the advertisement server 2 obtains the posted image is not particularly limited, and data necessary for the implementation of the present invention is provided from the operator of the information providing service via the Internet such as SNS. You may receive it. For example, the operator of the advertisement server 2 may purchase or provide raw data posted or an analyzed report from the operator of the SNS server 3. Further, the advertisement server 2 and the SNS server 3 cooperate with the system via, for example, an API (Application Programming Interface), and the advertisement server 2 acquires information necessary for determining the related image from the SNS server 3 as needed. May be.

  Therefore, in the present invention, in order to acquire images and posts by the advertisement server 2, the advertisement server 2 collects images and posts from the SNS server 3, and the advertisement server 2 collects the SNS server 3 and other servers. This includes receiving images and posts.

  In the above description, the related image determination unit 24 receives the designation of the search period and collects the posted images included in the posts published from the SNS server 3 within the designated period. However, the designation of the search period is not necessarily required. The related image determination unit 24 only needs to be able to acquire a plurality of posts to the SNS server 3 (whether published or unpublished) corresponding to a predetermined period.

  FIG. 11 is an explanatory diagram showing an example of a posted image DB in the information provision management method according to the present embodiment. As shown in FIG. 11, the posted image DB is configured such that, for example, a posted ID, posted text information, time stamp, location information, a poster ID of the poster, posted image ID, posted image data, and the like are registered. Yes.

<Extraction of Fifth Image Group B ′>
Next, the related image determination unit 24 extracts a posted image including the registered image from the fourth image group A ′ registered in the posted image DB, and stores it in the image group B ′ table as the fifth image group B ′. sign up.

  FIG. 12 is a schematic diagram showing a specific example of posted images belonging to the fifth image group B ′ in the information provision management method according to the present embodiment. As illustrated in FIG. 12, the posted image 31 included in the fifth image group B ′ includes an image that is the same as or similar to the registered image 32 as an object, and an image 33 of another object. ing. In the following description, for convenience, an image that is the same as or similar to the registered image 32 is also referred to as a registered image 32. Other subjects are people, products, logos, landmarks, landscapes, etc., and are not particularly limited. The posted image 31 included in the fifth image group B ′ includes an image showing only the registered image 32.

  FIG. 13 is a flowchart showing each step of the fifth image group B ′ extraction process in the information provision management method according to the present embodiment. As shown in FIG. 13, the related image determination unit 24 (see FIG. 4) reads the Nth (N initial value 1) posted image from the posted image DB (S31), and the posted image is the feature amount of the registered image. (S32).

  If Yes in S32, the posted image is registered in the image group B 'table (S33). On the other hand, if No in S32, the process proceeds to S34.

  After incrementing N by 1 in S34, it is determined whether there is an Nth registered image (S35). If yes in S35, the process returns to S31, and if No, the process is terminated.

  The configuration of the image group B ′ table is the same as that of the posted image DB described with reference to FIG.

<Extraction of Sixth Image Group C1, C2,... Ck>
Next, the related image determination unit 24 extracts images including candidate images from the posted images included in the fifth image group B ′ registered in the image group B ′ table, and the sixth image groups C1, C2,. ..Ck is obtained.

  FIG. 14 is an explanatory diagram showing an example of a candidate image DB in the information provision management method according to the present embodiment. The candidate image DB is configured such that, for example, a candidate image ID, a candidate image name, and candidate image data are registered.

  In the present embodiment, as described later, the postage image including the candidate image is determined using the feature amount of the candidate image. However, instead of the feature amount or together with the feature amount, the subject image includes the subject. Text data such as product names and shape data (for example, CAD data) such as products can also be used. In this case, the candidate image DB shown in FIG. 14 is further configured to register text data and subject shape data.

  FIG. 15 is a flowchart showing the steps of extracting the sixth image group C1, C2,... Ck in the information provision management method according to the present embodiment. As shown in FIG. 15, the related image determination unit 24 (see FIG. 4) first accepts selection of a candidate image (S41). Selection of candidate images can be performed semi-automatically. That is, the advertisement server 2 reads out a plurality of candidate images from the candidate image DB, displays them on the operator's terminal, and the operator selects from them (check box format or the like). However, the operator may manually select a candidate image from all candidate images.

  Next, the related image determination unit 24 reads the Kth (K initial value 1) candidate image from the plurality of selected candidate images, and calculates a feature amount (S42).

  Thereafter, the related image determination unit 24 reads the Nth (N initial value 1) posted image from the posted image DB (S43), and determines whether the posted image has the feature amount of the candidate image (S44). ).

  If Yes in S44, the posted image is registered in the image group Ck table (S45). On the other hand, if No in S44, the process proceeds to S46.

  After incrementing N by 1 in S46, it is determined whether there is an Nth posted image (S47). If Yes in S47, the process returns to S43 and repeats S43 to S47.

  If “No” in S47, the processing has been completed for all the posted images registered in the posted image DB using the Kth candidate image, and the process proceeds to S48. In S48, after K is incremented by 1, it is determined whether or not there is a Kth candidate image (S49). If Yes in S49, the process returns to S42 and S42 to S49 are repeated. If No in S49, the process ends.

  Through the processing as described above, an image group Ck table is created for all candidate images selected in S41.

  The configuration of the image group Ck table is the same as the posted image DB described with reference to FIG.

<Affinity image selection>
FIG. 16 is a schematic diagram showing a specific example of posted images belonging to the sixth image group C1, C2,... Ck in the information provision management method according to the present embodiment. The posted images 41 to 44 illustrated in FIGS. 16A to 16D include a registered image 32 and candidate images 45 to 48, respectively.

  It is assumed that the number n of posted images belonging to each of the sixth image groups C1, C2,... Ck is different. In the present embodiment, the compatibility between the registered image 32 and the candidate images 45 to 48 is obtained based on the number n, and the candidate image having high compatibility is called an affinity image and used as a related image.

  FIG. 17 is a flowchart showing steps of affinity image determination processing in the information provision management method according to the present embodiment. As illustrated in FIG. 17, the related image determination unit 24 (see FIG. 4) refers to the posted image DB and counts the total number N of posted images belonging to the fourth image group A ′ (S51). Next, the number n of posted images belonging to the image group Ck is counted with reference to the sixth image group Ck (the initial value of k is 1) (S52).

  Next, based on the total number N of posted images belonging to the fourth image group A ′ and the number n of posted images included in the sixth image group Ck, the related image determining unit 24 has affinity = n / N × 100 (%) is calculated (S53).

  Next, the related image determination unit 24 determines whether or not the calculated affinity is greater than or equal to a predetermined threshold (S54). If Yes in S54, the candidate image corresponding to the sixth image group Ck is determined as an affinity image (S55). If No in S54, the process proceeds to S56.

  In S56, the related image determination unit 24 increments k by 1, and then determines whether or not there is a next sixth image group Ck (S57). If Yes in S57, the process returns to S52, and if No, the process ends.

  In this way, the affinity is obtained for the sixth image group C1, C2,... Ck, and it is determined whether the candidate image is an affinity image.

  As described above, the related image determination unit 24 selects affinity images from candidate images and determines them as related images. The affinity image is used to create an AO list described below.

  As described in S21 of FIG. 10, the search period for collecting the fourth image group A ′ is specified because the affinity (compatibility) of the candidate images changes depending on the specified search period. is there. For example, in the case where the search period is the most recent three months and in the case where the search period is three months one year ago, the affinity may be different even for the same candidate image. By specifying the search period, it is possible to determine the posting images posted during a certain period of time that have a good affinity with the registered image, and to find the advertising opportunity, that is, the affinity opportunity, among those that do not include the registered image be able to.

  In the above description, the affinity is calculated for each of the sixth image groups C1, C2,... Ck based on the number n of posted images including candidate images and registered images. However, the content of the posting including the posting image including both the candidate image and the registered image is not necessarily favorable for the product, service, or brand related to the registered image.

  For example, as shown in FIGS. 7A and 7B, a registered image includes a product image of a soft drink “ABC soda”, and a candidate image includes a brand logo of a pizza “number one pizza”. In addition, regarding the pizza “number one pizza”, negative comments such as “poor” and “insufficient” may be included. Thus, even if there are many posts that are not favorable to the pizza “number one pizza”, the number n is increased and the affinity is increased. As a result, when the brand logo image of the pizza “No. 1 Pizza” is selected as an affinity image and an advertisement is placed on the target image, the effect of the advertisement is not only low, but on the contrary, the own brand is damaged. It is also possible.

  Therefore, the related image determination unit 24 has a predetermined number (a natural number) of negative words that are negative words that are not favorable to the product or service related to the candidate image in the posted text information for each post including the candidate image. ) Determine whether it is included or not. When a predetermined number or more of negative words are included, the post described in S52 of FIG. 17 is excluded from the count target, and the above number n is counted. Thereby, the viewpoint of whether a contribution is favorable or not is added to the determination of a related image, and the above-mentioned brand damage can be suppressed.

  In addition, as described above, the related image determination unit 24, when a predetermined number or more of negative words are included in a post, instead of excluding the post from the count target, the product or service related to the candidate image If the candidate image is determined as a related image when the number of unfavorable posts is equal to or greater than a predetermined threshold, the related image is counted separately as a non-favorable post. For, a post evaluation result including a related image indicating that there are many negative posts may be attached.

  Such an evaluation result of the posting is added to an advertisement target DB (see FIG. 22) described later by the AO list generation unit 25 of the advertisement server 2, and further created based on the advertisement target DB by the inventory management unit 26. Added to the inventory DB (see FIG. 23). Then, the submission management unit 22 controls the network control unit 102 to be included in the bid request transmitted to the advertiser terminal 6 via the network 7. As a result, the advertiser terminal 6 displays the evaluation result of the posting including the related image together with the image data of the target image, etc., so that the advertiser determines the candidate image in determining the candidate image with reference to the evaluation result. It is possible to grasp the fact that a plurality of posts belonging to each of the six image groups C1, C2, C3,.

  In consideration of brand damage as described above, advertisers avoid placing advertisements on related images with many negative posts, but in order to improve their brand for negative posts, You may want to distribute Thus, being able to refer to the evaluation result of a post including a related image at the time of bidding is beneficial in that efficient information provision can be performed.

  In the above description, a negative word is used, but the present invention is not limited to this. For example, the related image determination unit 24 does not include a positive word, which is a positive word indicating favor for a product or service, in the posted text information of a post including a candidate image, or is equal to or less than a predetermined number (natural number). The post may be excluded from the number n of the counting targets based on whether or not it is. In addition, the related image determination unit 24 separately counts the posting as a post that is not favorable for the product or service related to the candidate image, and the number of non-favorable posts for the candidate image is equal to or greater than a predetermined threshold. In addition, even when the candidate image is determined to be a related image, an evaluation result of a post including the related image indicating that there is no or a small number of positive posts may be attached to the related image.

  In addition, the related image determination unit 24 detects both positive words and negative words included in the post text information of posts including candidate images, and when a positive word is detected, the evaluation point P is incremented by 1, When a negative word is detected, the evaluation point P is decremented by one. Then, based on the final evaluation point P, the posting including the related image may be evaluated. For example, when the evaluation point P is −50 points or less, a post evaluation result including a related image indicating that there are many negative posts may be attached to the related image.

<Collection of first image group A>
In the present embodiment, a plurality of pre-release posts are collected from a plurality of licensed members, post images included in the plurality of posts are collected, and the first image group A is configured.

  FIG. 18 is a flowchart showing each process of the collection processing of the first image group A in the information provision management method according to the present embodiment. As illustrated in FIG. 18, the AO list generation unit 25 (see FIG. 4) of the advertisement server 2 receives a post before release (hereinafter referred to as a pre-post) from the member terminal 5 (S 61). Here, the advance posting is a posting before posting from the member terminal 5 to the SNS server 3. Next, the AO list generation unit 25 acquires a posted image included in the received prior posting (S62). In addition, the AO list generation unit 25 acquires post-related information included in the received pre-post (S63). Furthermore, the AO list generation unit 25 associates the collected posted images and posted image-related images and registers them in the pre-posted image DB (S64).

  The configuration of the pre-post image DB is the same as that of the registered image DB described with reference to FIG.

<Extraction of second image group B>
FIG. 19 is a flowchart showing each step of the second image group B extraction process in the information provision management method according to the present embodiment. As illustrated in FIG. 19, the AO list generation unit 25 (see FIG. 4) reads the Nth (N initial value 1) posted image from the previously posted image DB (S <b> 71), and the posted image is a feature of the registered image. It is determined whether or not there is an amount (S72).

  If Yes in S72, the posted image is registered in the image group B table (S73). On the other hand, if No in S72, the process proceeds to S74.

  After incrementing N by 1 in S74, it is determined whether there is an Nth posted image (S75). If the answer is Yes in S75, the process returns to S71, and if No, the process is terminated.

  The configuration of the image group B table is the same as the posted image DB described with reference to FIG.

<Extraction of third image group D>
Next, a posted image including an affinity image is extracted from the second image group B to obtain a third image group D.

  FIG. 20 is a flowchart showing each step of the third image group D extraction process in the information provision management method according to the present embodiment. As shown in FIG. 20, the AO list generation unit 25 (see FIG. 4) calculates the feature amount of the affinity image (S81). Next, the AO list generation unit 25 reads the Nth (N initial value 1) posted image from the image group B table (S82), and determines whether the posted image has an affinity image feature amount (S82). S83).

  If Yes in S83, the posted image is registered in the image group D table (S84). On the other hand, if No in S83, the process proceeds to S85.

  After incrementing N by 1 in S85, it is determined whether there is an Nth posted image (S86). If yes in S86, the process returns to S82, and if No, the process is terminated.

  Here, for convenience of explanation, the description has been made on the premise that there is one affinity image. However, if there are a plurality of affinity images, the steps S81 to S86 may be repeated accordingly.

  The configuration of the image group D table is the same as the posted image DB described with reference to FIG.

<Create AO list>
Next, the target image is extracted from the third image group D by excluding post images belonging to the third image group D and also belonging to the second image group, and an AO list is created.

  FIG. 21 is a flowchart showing each step of the affinity opportunity list creation process in the information provision management method according to the present embodiment. As shown in FIG. 21, the AO list generation unit 25 (see FIG. 4) reads out the image ID of the Mth (M initial value 1) posted image from the image group D table (S91), and stores it in the image group B table. It is determined whether or not there is (S92).

  If No in S92, the posted image and posted related information corresponding to the image ID are read from the image group D table (S93) and registered in the advertisement target DB (S94). On the other hand, if Yes in S92, the process proceeds to S95.

  After incrementing M by 1 in S95, it is determined whether or not there is an Mth posted image (S96). If yes in S96, the process returns to S91, and if no, the process ends.

  Through the above process, the advertising target information including the target image and the posting related information of the posting including the target image is registered in the advertising target DB. The data stored in this advertisement target DB is an AO list.

  FIG. 22 is an explanatory diagram showing an example of an advertisement target DB (affinity opportunity list) in the information provision management method according to the present embodiment. As shown in FIG. 22, the advertisement target DB is configured to register, for example, a posting ID, posting text information, time stamp, position information, a poster ID of a poster, a posting image ID, posting image data, and the like. Yes.

<Inventory management>
Next, the inventory management unit 26 adds the advertisement target DB (AO list) and the licensed member DB to generate an inventory DB.

  FIG. 23 is an explanatory diagram showing an example of an inventory DB in the information provision management method according to the present embodiment. As shown in FIG. 23, the stock DB includes, for example, a post ID, post text information, time stamp, position information, poster member ID, post image ID, post image data, post member name, and number of followers. , Member-related information such as the number of posts, follow-up member ID, etc., influence level of the poster (described later), minimum bid amount (described later), compatibility with registered image (described later), successful bidder ID (described later), and successful bid amount (Described later) is registered.

  FIG. 24 is a schematic diagram illustrating a specific example of a target image in the information provision management method according to the present embodiment. As shown in FIGS. 24A and 24B, the target image 51 includes a pizza affinity image 52 that is compatible with the soft drink registration image 32 shown in FIG. 7, but does not include the soft drink registration image 32.

<Bid>
In the present embodiment, a case will be described as an example where the present invention is applied to an operational advertisement and the publisher of the advertisement for the target image registered in the inventory DB is determined by bidding.

  FIG. 25 is a flowchart showing each step of the process for setting the minimum bid for an advertisement in the information provision management method according to the present embodiment. As shown in FIG. 25, the manuscript management unit 22 (see FIG. 4) obtains the influence level of the contributor on the Xth target image (the initial value of X is 1) from the inventory DB (S101).

  Specifically, as the influence level of the poster, for example, the number of posters to be disclosed (followers, etc.), the number of reposts (retweets, shares, etc.) for posts by the posters, and comments on posts by the posters , The number of responses (such as acceptance of “favorites”, “likes”, “likes”, etc.) to the posting by the poster, the number of impressions of the posting by the poster, and the number of reach.

  Here, the impression is that a post is displayed, and the number of impressions is the number of times the post is displayed. The reach number is a so-called number of unique users excluding those due to duplicates among impressions.

  More specifically, member-related information such as the number of followers and the number of posts included in the inventory DB shown in FIG. 23 can be used.

  Further, the influence level of the poster may be calculated based on the member related information as well as the member related information itself.

  Next, the placement management unit 22 determines the minimum bid amount of the target image based on the influence level of the poster (S102). For example, the minimum bid amount of other advertising media close to the influence level of the poster may be set as the minimum bid amount of the target image. More specifically, a medium DB in which the degree of influence of the advertising medium and the minimum bid amount are associated and registered is prepared, and the degree of influence of the poster acquired in S101 and the degree of influence of the advertising medium registered in the medium DB are obtained. In contrast, the same or closest advertising medium is selected, and the minimum bid amount attached to the advertising medium is set as the minimum bid amount of the target image.

  As another method, the number of PVs of a plurality of websites is investigated using a website automatic analysis program. In addition, the minimum bid amount for the advertisement to the website is obtained. By comparing the number of followers of the poster and the number of PV of the website, the website with the closest number of PV is selected, and the minimum bid of the advertisement to the website is determined as the minimum of the target images included in the posting by the poster The bid amount.

  If the minimum bid amount is determined in S102, the placement management unit 22 registers the minimum bid amount in the inventory DB (S103).

  Next, after incrementing X by 1 in S104, it is determined whether or not the Xth target image is in the inventory DB (S105). If Yes in S105, the process returns to S101, and if No, the process ends.

  By such processing, the minimum bid amount is determined for the target image registered in the inventory DB.

  FIG. 26 is a sequence diagram showing each step of an advertisement bidding process in the information provision management method according to the present embodiment. As shown in FIG. 26, the advertisement management unit 22 (see FIG. 4) of the advertisement server 2 transmits a bid request to a plurality of advertiser terminals 6 (see FIG. 1) (S111). The bid request is created with reference to the inventory DB. In the bid request, for example, in addition to information that can specify an advertisement placement target such as image data of the target image, for example, member-related information of a poster including the target image, a minimum bid amount (described later), and an influence degree Among them, at least one is included.

  The information included in the bid request is displayed on the advertiser terminal 6 on the display unit 203 (see FIG. 3), and serves as a reference for determining whether the advertiser needs to bid and determining the bid amount.

  Here, it is preferable that the manuscript management unit 22 calculates the degree of compatibility of the posted poster with the registered image. The degree of compatibility of the poster with the registered image refers to the degree of interest or favor of the product, service, company, etc. corresponding to the registered image by the poster. The information related to the degree of compatibility includes the degree of compatibility calculated as described later, and indirect information such as a determination result indicating that the degree of compatibility is high or low.

  Here, the case where the registered image includes a product image or brand logo of the soft drink “ABC soda” as shown in FIGS. 7A and 7B will be described as an example, and the degree of compatibility with the registered image will be described. When a certain contributor frequently refers to the soft drink “ABC soda” in the posting and there are many positive comments on “ABC soda”, it is considered that the degree of compatibility is high. In addition, when a plurality of soft drink brand products frequently appear in a posted image included in a post by a certain contributor, it is considered that the degree of compatibility with the soft drink “ABC soda” is low. In addition, when a certain poster frequently refers to a soft drink competing with “ABC soda”, or when there are many negative comments on “ABC soda”, the compatibility with the soft drink “ABC soda” is high. It is considered to be low and incompatible.

  Advertisers who want to place advertisements for products and brands that compete with “ABC Soda” will only be able to place ads when the poster is highly compatible with the registered images. In addition to this, it is also conceivable that our own brand will be damaged. It is expected that advertisers do not want to place advertisements for contributors who have low compatibility with registered images.

  In order to prevent such brand damage from occurring, the placement management unit 22 refers to the inventory DB shown in FIG. 23 and analyzes post text information and post image data of posts by the same contributor. The degree of compatibility with the registered image of the poster is calculated. And it is preferable that the advertisement management part 22 transmits the information regarding the calculated compatibility degree to the advertiser terminal 6 together with the advertisement target information and the price of information provision. Thereby, the advertiser can refer to the information related to the degree of compatibility with the registered image of the poster and determine whether or not to bid.

  The method for calculating the degree of compatibility with the registered image of the contributor is not particularly limited, and is performed as follows, for example. FIG. 27 is a flowchart showing each step of the calculation process of the degree of compatibility with the registered image of the poster in the information provision management method according to the present embodiment. As shown in FIG. 27, the manuscript management unit 22 (see FIG. 4) obtains post text information about the Xth target image (the initial value of X is 1) from the inventory DB (S111).

  Next, the manuscript management unit 22 counts the positive words included in the posted text information (S122). For example, a table in which a plurality of positive words are registered is prepared, and the number of positive words included in the table is counted in the posted text information.

  The value counted in S122 is added to the previous value (S123). Next, after incrementing X by 1 in S124, it is determined whether or not the Xth target image is in the inventory DB (S125). If Yes in S125, the process returns to S121.

  If it is No in S125, it is determined whether or not the total number of positive words included in the posting of the poster as the compatibility degree is equal to or greater than a predetermined threshold Y (S126). If Yes in S126, it is determined that the degree of compatibility with the registered image of the poster is high, and is registered in the posting DB (S127). If No in S126, S127 is skipped and the process is terminated.

  By performing such a process for each contributor, the degree of compatibility of all the contributors with respect to the registered image can be obtained.

  Here, the posted text information associated with the target image included in the inventory DB is used to calculate the degree of compatibility with the registered image. However, the manuscript management unit 22 controls the network control unit 202. The data of the past posts of the posters may be obtained separately from the SNS server 3 via the network 7 and used.

  The method of presenting information related to the degree of compatibility with the registered image of the poster in the advertiser terminal 6 is not particularly limited. For example, when the degree of compatibility with the registered image of the poster is high, it corresponds to posting by the registrant. The bid request can be displayed on the advertiser terminal 6 with a mark or the like indicating a warning (alert) to the advertiser who performs an advertisement of another product or brand.

  Although a positive word is used here, the present invention is not limited to this. For example, the manuscript management unit 22 may count negative words and determine that the degree of compatibility with the registered image of the poster is low when the number of negative words is equal to or less than a threshold value. In this case, in the advertiser terminal 6, a bid request corresponding to the posting by the registrant is accompanied by a mark or the like indicating a warning (alert) to the advertiser who performs a product or brand advertisement related to the registered image, Display.

  Also, both positive words and negative words included in the posted text information are detected. When a positive word is detected, the compatibility point P ′ is incremented by 1. When a negative word is detected, the compatibility level The point P ′ is decremented by 1. Then, the degree of compatibility with the registered image of the poster may be determined based on the final compatibility point P ′. For example, when the compatibility point P ′ is +50 points or more, it is determined that the compatibility level is high, and when the compatibility point P ′ is −50 points or less, the compatibility level is determined to be low. In this case, the advertiser terminal 6 marks the bid request corresponding to the posting by the registrant so that the advertiser can identify the case where the compatibility with the registered image of the poster is high and the case where it is low. Etc. are attached and displayed.

  Returning to FIG. 26, next, bid information is transmitted from the plurality of advertiser terminals 6 to the placement management unit 22 (S112). The bid information includes, for example, an advertiser ID and a bid amount.

  Next, based on the bid amount included in the received bid information, the placement management unit 22 determines, for example, the advertiser who has presented the highest bid amount as a successful bidder (S113). Further, when the successful bidder is determined, the manuscript management unit 22 registers the successful bidder information in the inventory DB (S114). The winning bidder information is, for example, the winning bidder ID that is the advertiser ID of the winning bidder or the winning bid amount that is the bid amount of the winning bidder.

  In the present embodiment, the case where the advertiser makes a bid directly using the advertiser terminal 6 has been described as an example. However, like the above-mentioned DSP, a bid agency server that performs bidding on behalf of the advertiser may perform bidding. In this case, the bid proxy server selects an advertiser that can present the highest bid amount among a plurality of advertisers, and transmits bid information to the placement management unit 22 on behalf of the advertiser.

<Advertisement delivery>
FIG. 28 is a sequence diagram showing each step of the advertisement distribution process in the information provision management method according to the present embodiment. As shown in FIG. 28, first, the advertisement providing unit 23 (see FIG. 4) of the advertisement server 2 transmits a post with advertisement (S131).

  FIG. 29 is a flowchart showing each step of the posting process with the advertisement in the information provision management method according to the present embodiment. As illustrated in FIG. 29, the advertisement providing unit 23 acquires advertisement creation information included in the advertiser DB illustrated in FIG. 6 (S151).

  The advertisement creation information (creative information) includes, for example, the type of advertisement, the image (product, logo, character, etc.) and text constituting the advertisement, and the layout information of the advertisement. Further, the advertisement creation information may include a URL of a website such as an advertiser.

  The advertisement type is, for example, (I) a post in which an advertisement is added to the target image and (II) a post in which an advertisement is added to the post. The advertisement layout information includes the position and size of the advertisement to be added.

  Next, the advertisement providing unit 23 creates a post with an advertisement based on the advertisement creation information (S152). FIG. 30 is a schematic diagram showing a specific example of an advertisement in the information provision management method according to the present embodiment. FIG. 30A shows an example in which the advertisement type is (I). An advertisement image 62 including a soft drink image and a tagline text is superimposed on the target image 61.

  FIG. 30B shows an example in which the advertisement type is the above (II). In the post with advertisement 63, an advertisement 66 including a tagline text and a logo indicating an advertiser is added to the original post 65 including the target image 64.

  When the advertisement creation information includes the URL of the website, the advertisement providing unit 23 embeds a link to the website in the post with advertisement.

  Returning to FIG. 29, the advertisement providing unit 23 transmits a post with advertisement to the SNS server (S153). At this time, the advertisement providing unit 23 reads the member ID and password from the licensed member DB shown in FIG. 9 and posts a post with an advertisement to the SNS server 3 on behalf of the licensed member.

  Returning to FIG. 28, the SNS server 3 displays the posting with advertisement on the member terminal 5 (S132), reposts (retweet, etc.) (S135), clicks a link to the website included in the posting with advertisement (S138), and the like. When any action on the posting with advertisement by the member is detected, the advertisement providing unit 23 is notified (S133, S136, S139). The advertisement providing unit 23 generates advertisement rewards (S134, S137, S140) in response to the notifications (S133, S136, S139). Thereafter, the advertisement providing unit 23 requests the settlement server 4 to perform a billing process for the generated advertisement based on the settlement related information acquired from the advertiser DB shown in FIG. 6 (S141). Further, the advertisement providing unit 23 calculates the affiliate consideration of the licensed member based on the advertisement reward amount, and pays the affiliate consideration based on the payment related information acquired from the licensed member DB shown in FIG. (S142). The settlement server 4 bills the advertiser for advertising rewards and pays affiliate fees to the licensed members.

  In addition to the case described above, the reward may be generated based on the transmission of the post with advertisement, and is not particularly limited. In addition to the case described above, the above-described action includes comments and reverberations (for example, called “like”) and the like to a post with advertisement, and is not particularly limited.

<Effect>
As described above, according to the information provision management method according to the present embodiment, a related image related to a registered image is obtained from the first image group A composed of a plurality of posted images included in a plurality of posts. Since the posting image that includes the registration image but does not include the registration image is extracted as the target image, and the advertising target information associated with the posting related information related to the posting including the target image is obtained, the registration image based on the advertising target information is obtained. Since related advertisements can be transmitted effectively, it is possible to effectively provide advertisements suitable for images included in posts.

  In the present embodiment, the method further includes a step of determining a related image from a plurality of candidate images, and is a parameter indicating the relationship between the candidate image and the registered image, which is calculated based on at least one of the number α. A related image is determined from a plurality of candidate images based on the affinity degree. Thereby, since a related image can be selected from candidate images based on the relevance with a registered image, more effective advertisement distribution can be made to SNS members.

  In this embodiment, since the minimum bid amount of the advertisement, which is one aspect of the price of providing information to the posting, is set based on the influence degree of the poster who made the posting, By setting the minimum bid amount corresponding to the value of the advertisement, it is possible to suppress the bid amount from becoming excessively high or low.

  As described above, the price for providing information is not limited to a minimum price such as a minimum bid amount, and may be a maximum price, an average price, and a reference price. Therefore, in the present embodiment, the case where bidding is started from the minimum bid amount set based on the influence level of the poster has been described as an example, but there is no minimum bid amount, that is, the minimum bid amount is zero. It is possible to apply the present invention even in the case of. In this case, the price of the advertisement calculated based on the influence of the poster is used as a reference price for the advertiser to determine the bid amount, and the price of the successful bid may be excessively high or low. Can be suppressed.

  In addition, the price of the advertisement calculated based on the influence level of the poster can be used for purposes other than bidding (request for information provision). For example, the advertisement providing unit 23 transmits a post with advertisement to the SNS server 3, and after the post with advertisement is distributed by the SNS server 3, the effect on the advertisement (hereinafter referred to as advertisement effect) is measured. The advertisement providing unit 23 calculates to convert the measured advertisement effect into an amount (hereinafter referred to as an advertisement conversion value). The advertisement providing unit 23 generates a report including the advertisement conversion value, the winning bid amount, and the price of the advertisement calculated based on the influence of the poster, and controls the network control unit 102 to control the network 7. Via the transmission to the advertiser terminal 6. Thereby, the advertiser can examine the cost effectiveness of the advertisement by comparing the advertisement price calculated based on the advertisement conversion value, the winning bid amount, and the influence degree of the poster. At this time, it is not necessary to include the price of the advertisement in the report. The cost of the advertisement calculated based on the advertisement conversion value, the winning bid amount, and the price of the advertisement calculated based on the influence of the poster is not required. Only the evaluation result of the effect (for example, “high advertising effect”, “A rank”, or “high advertising effect”) may be included in the report. From such an evaluation result, the advertiser can grasp the cost-effectiveness of the advertisement and can refer to it in the next advertisement placement.

  Here, the price of the advertisement, which is one aspect of the information provision price of the present invention, is controlled by the network management unit 102 by the placement management unit 22 or the advertisement provision unit 23, and the advertiser terminal 6 via the network 7. Has been sent to. However, the manuscript management unit 22 or the advertisement providing unit 23 may control an output device such as a display unit (not shown) or a printer unit (not shown) to output the price of the advertisement.

  In addition, the advertising effect includes, for example, the number of reposts (retweet, share, etc.) for posts with advertisements, the number of responses to posts by posters (registration to favorites, acceptance of likes, likes, etc.), advertisements The number of impressions or reach of post with posting, and the number of conversions or conversion rate by posting with advertisement.

  Further, in the present embodiment, pre-posts are collected from the licensed members and set as the first image group A. Therefore, it is possible to post to the SNS server 3 after adding an advertisement to the pre-posts. The licensed member is paid a part of the advertising reward as an affiliate. This increases the incentive for members to post images on the SNS and diffuse information, further increasing the advertising opportunities.

  In addition, this invention is not limited to the said embodiment, It can implement variously. In the above-described embodiment, the constituent elements illustrated in the accompanying drawings are not limited to this, and can be appropriately changed within a range where the effects of the present invention are exhibited. In addition, various modifications can be made without departing from the scope of the object of the present invention.

  In the above embodiment, an advertisement is described as an example of information provision. However, the present invention can be applied to provision of content and provision of PR in addition to provision of advertisement.

  In the above-described embodiment, advertisements are solicited from the target images listed in the AO list that are included in the posting of the licensed member, but are not limited to the licensed member, but are listed in the AO list. All target images may be in stock.

  In the above-described embodiment, the case where the minimum bid amount is determined based on the influence degree of the contributor who made the advance posting is described as an example. However, when placing an advertisement on a post that has already been published, the degree of influence of a post that has already been published may be used to determine the minimum bid amount. The degree of influence here is, for example, the number of comments on the post, the number of retweets, and the number of reflections.

  In the above embodiment, the posted image is extracted from the original database such as the posted image DB and registered in another database or table such as the fourth image group B ′ table, but extracted to the original database. The extracted post image may be managed by setting a flag indicating that the post image has been posted.

  In the above embodiment, the case where the present invention is applied to the SNS is mainly described as an example. However, the present invention is not limited to the SNS, and can be widely applied to services that provide information on the Internet. For example, the present invention can also be applied to a software infrastructure such as CMS (Contents Management System) for building a site such as a blog site or a member site.

  In the above embodiment, as described in (A) generation of advertisement target information, (B) determination of related images, and (C) inventory management, related images related to registered images are obtained from a plurality of posted images. Extract target images that include registered images but do not include registered images, and bid and distribute advertisements for posts that include target images, as described in (D) Managing advertisement placement and (E) Managing advertisement distribution ing. However, the bidding and distribution of advertisements described in (D) and (E) can be similarly applied to advertisement targets prepared by methods other than those described in (A) to (C).

  For example, the central control unit 101 of the advertisement server 2 controls the network control unit 102, collects a plurality of posts from the SNS server 3 via the network 7, and selects keywords satisfying a certain condition from these. It is also possible to extract a plurality of posts including the advertisement.

  Therefore, in the advertisement bidding and distribution described in (D) and (E) of the present embodiment, the posting to be advertised is not limited to the posting including the image, and the posting not including the image is also targeted. Can do.

  Therefore, another embodiment of the present invention is executed by a computer device including a network control unit, a storage unit, and a central control unit connected to a network, and the central control unit controls the network control unit. Receiving the information provision request via the network, and the central control unit controls the network control unit, and based on the information provision request, information about the posting Providing via a network, wherein the central control unit sets the price of information provision to the post based on the influence of the poster who made the post Information provision management method.

  Another embodiment of the present invention includes a network control unit, a storage unit, and a central control unit connected to a network, and the central control unit controls the network control unit to send an information provision request. Receiving via the network, controlling the network control unit to provide information targeted for the posting via the network based on the information provision request, and further, the central control The information provision management device, wherein the section sets a price of information provision to the posting based on an influence degree of a poster who has made the posting.

  With this configuration, the information provision requester (for example, the advertiser) refers to the information provision price calculated based on the influence degree of the poster, and provides information such as the necessity of information provision and the bid amount of the advertisement. Since it is possible to determine the amount of request for provision and to examine the cost-effectiveness of providing information, it is possible to provide information for posts that are cost-effective.

  According to the information provision management method and the information provision management apparatus of the present invention, there is an effect that information suitable for an image included in a post can be effectively provided, and in particular, it can be suitably used for advertisement distribution in SNS. .

DESCRIPTION OF SYMBOLS 1 Information provision management system 2 Advertising server 3 SNS server 4 Payment server 5 Member terminal 6 Advertiser terminal 7 Network 21 Advertisement object management part 22 Manufacture management part 23 Advertisement provision part 24 Related image determination part 25 AO list generation part 26 Inventory management part 100 Server 101, 201 Central control unit 102, 202 Network control unit 103, 207 Storage device 104, 208 Memory 105, 209 Power supply unit 200 Terminal 203 Display unit 204 UI device 205 GPS reception unit 206 Camera unit

Claims (8)

A computer device comprising a network control unit, a storage unit and a central control unit connected to a network executes,
The central control unit includes a related image associated with a registered image stored in advance in the storage unit from the image group A configured by a plurality of posted images included in a plurality of posts, and the registered image is included. Extracting the post image not present as a target image;
The central control unit associating and storing the target image and post-related information related to the post including the target image as target information in the storage unit;
An information provision management method comprising: Further comprising determining the related image from a plurality of candidate images;
Based on the image group A ′ composed of a plurality of posted images included in a plurality of posts on the network, the central control unit adds the number of posted images including each of the plurality of candidate images to the registered images. Calculated based on at least one of the number of posts containing the corresponding word, the number of posts containing a word indicating emotion, and the appearance probability of other words related to the specific word contained in the post 2. The information provision management method according to claim 1, wherein the related image is determined from the plurality of candidate images based on an affinity that is a parameter indicating a relationship between the candidate image and the registered image. .   The said central control part further comprises the process of providing the information which made the object image object the object via the said network by controlling the said network control part, The Claim 1 or Claim 2 characterized by the above-mentioned. Information provision management method described in 1.   3. The information provision management method according to claim 1, further comprising a step in which the central control unit sets a price of information provision to the target image based on the target information.   The information provision management method according to claim 4, wherein the setting of the price is performed based on an influence degree of a contributor of the posting.   The information provision management method according to claim 4, further comprising a step of calculating a degree of compatibility of the posting contributor with respect to the registered image by the central control unit.   7. The information provision management method according to claim 1, wherein the postings constituting the image group A are collected before posting. A network control unit connected to the network, a storage unit and a central control unit,
The central control unit is
From the image group A composed of a plurality of posted images included in a plurality of posts, including the related images related to the registered images stored in advance in the storage unit, and not including the registered images Extracted as an image,
The information provision management device, wherein the target image and post-related information related to the post including the target image are associated and stored as target information in the storage unit.

Images