Advertiser search system, advertiser search method, and program

Abstract

To provide a advertiser search system, a advertiser search method and a program for easily finding an advertiser suitable for advertising the goods and services of a client of advertisement. [Solution] The advertiser search system 100, which is connected to the SNS server and the proposal system via a network, comprises an SNS information acquisition unit 108 that acquires SNS account information of multiple candidates who could become advertisers and SNS post information posted using the SNS account information, and an information search unit 106 that searches for SNS account information that meets the conditions specified by the client from the SNS account information and SNS post information acquired by the SNS information acquisition unit 108.

Description

[Technical Field]

[0001] The present invention relates to a advertiser search system, an advertiser search method, and a program. [Background technology]

[0002] In recent years, it has become common for people to advertise their own products and services using media such as SNS (Social Network Service). To promote their products and services efficiently, it is also known to hire advertisers to promote their products and services on the advertisers' SNS. Among these advertisers are so-called influencers, who have influence (i.e., advertising power) over a specific customer segment (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2024-117038 Summary of the Invention [Problem to be solved by the invention]

[0004] The present disclosure aims to provide a system that makes it easy to find advertisers suitable for advertising products and services of a client requesting advertising. [Means for solving the problem]

[0005] In order to solve the above problem, one aspect of the present disclosure is a advertiser search system for a client to search for an appropriate advertiser, which includes an SNS information acquisition unit that acquires SNS account information of multiple candidates who could become advertisers and SNS post information posted using the SNS account information, and an information search unit that searches for SNS account information that meets conditions specified by the client from the SNS account information and SNS post information acquired by the SNS information acquisition unit. [Brief explanation of the drawings]

[0006] [Figure 1] FIG. 1 is a block diagram showing an outline of the entire system including the advertiser search system according to this embodiment. [Figure 2] FIG. 2 is a block diagram showing an outline of the advertiser search system according to this embodiment. [Figure 3] FIG. 3 is a schematic diagram illustrating an example of user information of a candidate. [Figure 4] Figure 4 shows an example of a database listing candidates. [Figure 5] FIG. 5 is a schematic diagram showing the structure of the SNS account information database. [Figure 6] FIG. 6 is a schematic diagram showing the structure of the SNS posting information database. [Figure 7] FIG. 7 is a schematic diagram showing the structure of the client information database. [Figure 8] FIG. 8 is a schematic diagram showing the structure of the competitive customer information database. [Figure 9] FIG. 9 is a schematic diagram showing the structure of a direct mail (DM) information database. [Figure 10] FIG. 10 is a schematic diagram showing the structure of the candidate list (search result) database. [Figure 11] FIG. 11 is a flowchart showing the overall processing flow of the advertiser search system in this embodiment. [Figure 12] FIG. 12 is a flowchart showing the overall processing flow of the advertiser search system in this embodiment. [Figure 13] FIG. 13 is a flowchart showing the scraping process flow in the SNS information acquisition unit. DETAILED DESCRIPTION OF THE INVENTION

[0007] The following describes embodiments of the present disclosure. The advertiser search system and its components according to the embodiments are merely illustrative, and each configuration can be modified as appropriate without departing from the spirit of the present disclosure. In particular, in the following embodiments, the hardware and functional blocks for realizing the terminals that make up the advertiser search system are clearly explained in detail. However, it is clear that which hardware realizes what function can be changed as appropriate within the spirit of the present disclosure. It is also clear that each functional block can be realized by the same hardware or software, or by different hardware or software. In the following description, it is assumed that an application (program) installed on the terminal controls each piece of hardware.

[0008] The advertiser search system according to the embodiment is a system that extracts advertisers suitable for advertising the products, etc. of a client (a person who intends to search for advertisers using the advertiser search system) that provides some kind of product or service (hereinafter, sometimes referred to as "products, etc.") and presents them to the client. In the following example, the advertiser search system issues a command to a proposal system using a machine learning model to search for advertisers, and presents the advertisers suggested in response to the command to the client. In short, the advertiser search system is a system that uses so-called AI (artificial intelligence) to search for the most suitable advertiser.

[0009] 1 is a block diagram showing an outline of a advertiser search system and its peripheral systems. A advertiser search system 100 is connected to an SNS server 400 and a proposal system 300 via a network 200.

[0010] (Advertiser Search System) As shown in Figure 2, the advertiser search system 100 can take various forms, such as an application installed on a client's personal computer or a browser application running on a browser. The advertiser search system 100 comprises an information acquisition unit 102, an information transmission unit 104, an information search unit 106, and an output unit 118. Each unit constituting the advertiser search system 100 is driven using the hardware of the personal computer based on commands from the advertiser search program.

[0011] (Information acquisition department) The information acquisition unit 102 mainly executes various processes for acquiring information from the outside. Specifically, the information acquisition unit 102 includes an SNS information acquisition unit 108, an email acquisition unit 110, and a customer information acquisition unit 112.

[0012] (SNS information acquisition department) The SNS information acquisition unit 108 accesses the SNS server 400 and acquires SNS information and the like uploaded to the SNS server 400. Specifically, the SNS information acquisition unit 108 acquires SNS account information of multiple candidates who can become advertisers, and SNS posted information posted using the SNS account information. Potential advertisers are those who meet predetermined conditions among the many users who upload information to the SNS server 400 (preferably excluding the requester).

[0013] (SNS account information) SNS account information is user information of SNS users, and includes various information such as the account holder's name, related area, related categories, number of followers, average engagement, gender, age, PR rate, average number of likes, number of engagements, gender ratio, and followers (including followers who follow those followers).

[0014] As shown in Figure 5, the SNS account information database includes items such as user name, profile information, related areas, related categories, follower information, and SNS account evaluation indicators (Figure). More details are as follows: The username is the name of the account holder, and may be, for example, a real name or a handle name.

[0015] The profile information includes, for example, information such as age and gender.

[0016] Age refers to the account holder's age or age range (e.g., 20s, 30s, 40s, etc.).

[0017] Gender refers to the sex of the account holder (e.g., male, female, other).

[0018] Related areas refer to areas where postings and promotional activities are primarily conducted, such as prefectures, cities, towns, villages, specific commercial facilities, and tourist destinations.

[0019] Related categories include, for example, beauty, cosmetics, personal care, restaurants, cafes, and the like.

[0020] The follower information includes, for example, the number of followers, the gender ratio (gender distribution of followers), and account information of followers and followers who follow those followers.

[0021] The number of followers refers to the number of users who follow a particular social media account and view it regularly.

[0022] The gender ratio indicates the ratio of male to female followers of a particular social media account, and is an indicator of the gender distribution of followers. The gender ratio is used when selecting advertisers whose follower attributes match the target demographic of the product or service being promoted. The gender ratio is calculated using the following formula: Gender ratio (%) = (Number of followers of a specific gender / Total number of followers) x 100 A follower is a user who follows a specific social media account and regularly views its posts (social media account information).Followers also include the user information (social media account information) of followers who follow the follower.

[0023] Social media account evaluation indicators include average engagement (average value of likes, comments, shares, etc.), average number of likes, PR rate (percentage of advertising posts), and number of engagements (total number of responses).

[0024] Average engagement refers to the total number of reactions (likes, comments, shares, saves, etc.) made by users to each social media post, averaged over a certain period or number of posts. Average engagement is calculated using the following formula: Average engagement = Number of posts in the target period Sum of engagements for each post in the target period / Number of posts in the target period The average number of likes refers to the average number of likes given to each post over a certain period or number of posts. The average number of likes is calculated using the following formula: Average number of likes = total number of likes for each post in the target period / number of posts in the target period The PR rate is an indicator that shows the percentage of advertising and promotional (PR) posts among all posts on a social media account. The PR rate is calculated using the following formula: PR rate (%) = (Number of PR posts / Total number of posts) x 100 Engagement refers to the total number of likes, comments, shares, saves, and other reactions users have made to a particular post. Engagement is calculated using the following formula: Engagement = Likes + Comments + Shares + Saves

[0025] (SNS posting information) SNS posting information refers to information including articles, photos, videos, audio, location information, hashtags, filters, effects, etc. posted to the SNS server 400. The SNS posting information database is a database that manages the content posted by candidates to SNS, and includes the following items. As shown in Figure 6, the SNS posting information database includes the posted content, post metadata, post accompanying information, etc.

[0026] Post content includes information such as articles (text), images (photos), videos, and audio. The post metadata includes information such as poster information (Author Data), posting date and time (Timestamp), posting location information (Geotag), posting language (Language), and post visibility settings (public, restricted, etc.).

[0027] Author Data is information about the account that created the post, including the account ID, username, profile picture, and verification badge (such as a blue badge indicating an official account).

[0028] The timestamp is the date and time (year, month, day, time, etc.) when a post was uploaded. This is used to analyze the frequency and timing of posts.

[0029] Post location information (Geotag) refers to the location where the post was made (e.g., restaurant, tourist spot, event venue). Location information makes it possible to analyze regions and areas where advertising is effective.

[0030] Posting language refers to the language in which the post is written (Japanese, English, Chinese, Korean, etc.), and is used to understand the post content, identify the target audience, and develop a promotional strategy.

[0031] Post visibility settings are information about the visibility of the post, and refer to settings such as "public," "private," "visible only to followers," and "visible only to specific users," which determine who can view the post.

[0032] Post-related information includes hashtags (keywords for classifying post content) and filter / effect information (information about image and video editing).

[0033] Hashtags are keywords or phrases attached to posts that categorize or relate to the content of a post. Hashtag analysis can help understand the subject matter of a post and the interests of its target audience.

[0034] Filter / Effect Data is information that indicates the types of filters and effects applied to posted images, videos, etc. This makes it possible to analyze the poster's preferred style of expression and preferences.

[0035] The SNS information acquisition unit 108 acquires the necessary information and searches for users who meet predetermined conditions from among a large number of SNS users through so-called scraping or crawling. The SNS information acquisition unit 108 also processes the acquired information and converts it into a format suitable for the UI screen of the advertiser search program.

[0036] The "predetermined conditions" used by the SNS information acquisition unit 108 are conditions related to the requester. The conditions related to the requester are conditions specified in advance by the requester or conditions particularly related to the requester's products, etc. For example, if the requester runs a coffee shop, conditions particularly related to the requester's products are expected to be conditions such as coffee shop, confectionery, and restaurant. The conditions related to the requester use store information (requester's store information), which will be described later. Furthermore, the conditions that the requester can specify may include information about the advertiser's profile.

[0037] The advertiser's profile includes, for example, the person's relevant area of ​​residence and area of ​​activity, particularly the relevant categories in which they post on social media (articles, photos, location information, hashtags, etc.), the number of followers they have, the followers they follow, etc., as well as information obtained from other social media account information. In other words, depending on how you specify certain conditions, you can search for so-called influencers who are best suited to promoting so-called products, etc.

[0038] (Email acquisition section) The email acquisition unit 110 acquires email information relating to the content of direct mail from multiple candidates who can become advertisers. The email acquisition unit 110 acquires email information relating to the content of emails sent and received by the candidates, which is stored in the SNS server 400.

[0039] The emails sent and received by the candidate are information about the contents of emails and direct messages that the candidate has sent and received in the past to other SNS users.

[0040] The email information is stored in a direct mail (DM) information database. As shown in Figure 9, the direct mail (DM) information database is a database that manages the sending and receiving history of direct mail within the SNS of potential advertisers, and includes email sending and receiving information, past advertising request information, etc. Specifically, it is as follows:

[0041] The email sending / receiving information includes the sending / receiving date and time, sender and destination information (email address, account ID, etc.), email body content, and attachment information.

[0042] Past advertising request information includes the details of the product or service being advertised, the advertising method (image posting, text posting, etc.), compensation conditions (amount of compensation, payment method, etc.), and the candidate's response status (whether or not they responded, the time required to respond, etc.).

[0043] The email acquisition unit 110 may acquire information by directly accessing the SNS server 400 using, for example, an API (Application Programming Interface), or may issue a command to the proposal system 300 to acquire emails of candidates who could become advertisers. In other words, it is also possible to acquire the information indirectly via the proposal system 300. When the information is acquired via the proposal system 300, a machine learning model can filter and organize the email information in advance, and provide it as information evaluating the response tendency and reliability of each candidate to advertising requests.

[0044] As will be explained in more detail later, by selecting the most suitable advertiser taking into consideration the content of the emails sent and received by the candidate, it is possible to narrow down the candidates who are likely to be able to advertise in a manner that is closest to the requested advertising style.

[0045] The acquired email information is then used by the information search unit 106 and the command sending unit 114 as an important indicator when selecting candidates, and is utilized as a basis for selecting candidates who will have a high advertising effect or who are likely to be able to negotiate smoothly with the client.

[0046] (Customer Information Acquisition Department) The customer information acquisition unit 112 acquires customer information of SNS users who handle products that compete with the client's product, etc. Specifically, the customer information acquisition unit 112 refers to the client's user information to acquire information on SNS users who handle products that compete with the client's product, etc. Whether a user handles a product that competes with the client's product, etc. can be determined by taking into account the content of the product, etc., the area in which each user is active, the demographics of users of the product, etc. Next, the customer information acquisition unit 112 acquires user information on the customers (e.g., SNS followers) of the acquired SNS user. As will be described in detail later, by selecting the most suitable advertiser by taking into account the customer information, it is possible to narrow down the candidates who have influence over the customer. The customer information of SNS users who handle products that compete with the client's product, etc. is stored in a competing customer information database.

[0047] As shown in Figure 8, the competing customer information database includes items such as competing company (store) information and competing customer information. The competing customer information database is a database that manages information on customers (followers) of other parties that compete with the client's products and services, and includes items such as competing company (store) information and competing customer information.

[0048] Competitor company (store) information includes information such as competitor company names, competitor store names, and product / service information (category, description, target demographic, etc.).

[0049] Competitive customer information includes information such as the customer's (follower's) SNS account information (user name, attributes, number of followers, etc.).

[0050] (Information transmission section) The information transmitting unit 104 mainly performs various processes for transmitting information to the outside. Specifically, the information transmitting unit 104 includes a command transmitting unit 114 and a DM transmitting unit .

[0051] (Command transmission unit) The command sending unit 114 sends a command to the proposal system 300 to search for an appropriate advertiser taking into account the various information acquired by the information acquisition unit 102. As an example, the command sending unit 114 writes a command to search for the most suitable advertiser from among multiple candidates in the prompt part of the command, and sends the command to the proposal system 300 together with the various acquired information.

[0052] In another embodiment, the command sending unit 114 can analyze the contents of the email information acquired by the email acquisition unit 110 as well as other information including store information and candidate profile information, and send a command to the proposal system 300 to rank multiple candidates who could become advertisers.

[0053] Store information (requester information) is basic information about the store of the requester registered in the advertiser search system 100, and specifically includes the store name, location (address), business hours, details of the products and services offered, contact information, homepage URL, store category, store photo and video information, word-of-mouth information, description (description of the store and services), payment method, availability of parking, and other various information registered or entered as information about the store. This information is stored in the requester information database. As shown in FIG. 7, the requester information database is a database that manages information about requesters who use the advertiser search system, and includes items such as basic information about the requester, information about the products and services offered, and store information.

[0054] The basic information of the requester includes information such as the store name or company name, location (address), contact information, and homepage URL.

[0055] The provided product / service information includes information such as the product or service name, product / service category, and product / service description (text, image, video, etc.).

[0056] Store information includes detailed information such as business hours, closing days, store photos, videos, reviews, payment methods, and whether parking is available.

[0057] Advertiser information refers to the advertiser's SNS account information, and includes various information such as the account holder's name (real name, handle name, etc.), related area, related category, number of followers, average engagement, gender, age, PR rate, average number of likes and engagements, male / female ratio, and followers (including followers who follow those followers).

[0058] (Command transmission unit) The command sending unit 114 can effectively extract and prioritize candidates that match the characteristics and target demographic of the client's store by analyzing the store information and advertiser information together with the email information. This makes it possible to accurately select advertisers who will maximize the advertising effectiveness of the client.

[0059] (DM sending department) The direct mail sending unit 116 sends a direct mail requesting advertisement to the candidate selected by the client in response to the proposal, in accordance with an instruction from the client.

[0060] (Information Search Department) The information search unit 106 searches for SNS account information that meets the conditions specified by the requester from the acquired SNS account information and SNS post information. Specifically, the information search unit 106 executes a series of operations to generate a command that takes into account the acquired information and the requester's wishes. The command generated here is sent to the proposal system 300 via the command sending unit 114.

[0061] (output section) The output unit 118 displays text and images (including still images and moving images), and generates information to be displayed on a display device such as a CRT display, a liquid crystal display, an organic EL display, a plasma display, or a touch display.

[0062] (network) The network 200 includes, for example, the Internet, a WAN (Wide Area Network), a LAN (Local Area Network), a provider terminal, a wireless communication network, a wireless base station, a dedicated line, etc. It is not necessary that all combinations of the devices shown in Fig. 1 are able to communicate with each other, and the network N may partially include a local network.

[0063] (SNS server) The SNS server 400 is a server that is configured to be accessible by a large number of users and is operated by a provider of an SNS service. For the sake of convenience, in this embodiment, it is assumed that only one SNS server is accessed to search for a advertiser, but multiple SNS servers operated by multiple SNS service providers may be accessed to search for a advertiser.

[0064] (Proposed system) The proposal system 300 is a device or system that utilizes AI (Artificial Intelligence). The generation AI uses a language model that has learned from information sites, including numerous information posting sites, and outputs an answer to an input request (question). The proposal system 300 can utilize, for example, a device or system that implements a tool using a machine learning model, such as a large-scale language model (LLM), such as a publicly known Chat GPT, Gemini, or Claude. In an embodiment of the present disclosure, the machine learning model is one that has been trained to output computer-generated text based on text information included in received communications. Note that the machine learning model is not limited to a large-scale language model, and may be any model that can be automatically generated by AI, and is not particularly limited thereto.

[0065] The proposal system 300 selects candidates based on the command received from the advertiser search system 100 and transmits (proposes) the results to the advertiser search system 100. The output of the proposal system 300 may be one candidate or multiple candidates. If there are multiple candidates, the proposal system 300 transmits a list of the candidates to the advertiser search system 100. At that time, it is preferable to include the user information of the candidates in the list of candidates.

[0066] In this embodiment, the proposal system 300 is an external system connected via the network 200, but it may also be a unique generation AI provided within the advertiser search system.

[0067] 2 is a schematic diagram showing an example of a client's user information. As shown in FIG. 2, the user information includes information such as the name of the user's product, multiple levels of business categories, and address. As described above, this information is used to search for SNS users who compete with the client's product. Specifically, the advertiser search system 100 can consider SNS users who offer products with the same or similar product names or business categories as competitors.

[0068] By taking into consideration the information acquired by the information acquisition unit 102 and requesting the advertiser to make a proposal from the proposal system 300, the following effects can be expected.

[0069] First, by taking into account the conditions related to the requester (the requester's social media account information) and the candidate's social media account information, it is possible to expect suggestions of candidates who are active in areas close to the requester's area of ​​activity. This, for example, increases the likelihood that the advertiser will accept the request and reduces advertising costs (expenses, etc.). Furthermore, by taking into account information on the social media account information, such as PR rate, average number of likes, and number of engagements, it is possible to expect suggestions of highly influential candidates.

[0070] Furthermore, by taking into account the information posted on social media by candidates, it is possible to expect the proposal of candidates who advertise in a style similar to the client's desired advertising style. For example, if the client wishes to advertise using a lot of photos, the client can include this preference in the proposal conditions and send a command to the proposal system 300. In response, the proposal system 300 can analyze the information posted on social media, select candidates who use a lot of photos, and propose them to the client.

[0071] Furthermore, by taking into account the content of emails sent by candidates, it is possible to expect proposals that take into account how the candidate responded to past advertising requests. For example, if the candidate has requested compensation for each past advertising request, it is possible to know the details of that request. Here, if the requester sends an instruction with the condition, for example, "candidates who will advertise for free," it is possible to expect proposals that exclude candidates who have requested compensation for advertising in the past.

[0072] In this way, according to the present invention, candidates who meet the advertising conditions desired by the client can be efficiently extracted, thereby eliminating the cumbersome work of researching influencers individually as in the past, and maximizing the advertising effectiveness of the client's products and services.

[0073] The list of candidates received from the proposal system 300 is stored and managed as a database by the advertiser search system 100.

[0074] Figure 3 is an example of a database showing a list of candidates. The list of candidates includes various user information such as the candidate's username, profile, number of followers, male-to-female ratio of followers, gender, average number of likes, etc. This information is managed in a format that can be narrowed down by filtering. As an example, by filtering the database to specify only those with a certain number of followers, it is possible to display only candidates with a large number of followers. This allows the selection of candidates that meet the client's wishes.

[0075] Alternatively, for example, by selecting candidates by checking the checkboxes in the leftmost column of the database and performing a predetermined operation, direct mail requesting advertising may be sent only to the selected candidates.

[0076] Next, a series of operations of the advertiser search system will be described.

[0077] As shown in Figure 11, the client first starts the advertiser search system. The operation of the advertiser search system 100 will be explained below, but the series of operations is performed by operating the client's personal computer under the client's operation.

[0078] The advertiser search system 100 acquires (scrapes) candidate information from the SNS server 400 (S101). At this stage, information on all candidates may be acquired, or information on selected candidates may be acquired after adding some conditions (for example, area of ​​activity, etc.).

[0079] Next, the advertiser search system 100 presents the candidate information to the client (outputs it to the client's personal computer) (S102). In response to this, the client considers the conditions for narrowing down the search. The client can specify, for example, the area in which the advertiser is active. The conditions specified by the client are input to the advertiser search system 100 in text format, for example. The conditions specified by the client may be selectable using check boxes, etc.

[0080] Next, the information search unit 106 sends a command to the proposal system 300 to narrow down the candidate list based on the conditions specified by the requester (S103).

[0081] The advertiser search system 100 then displays the candidate list received from the proposal system 300 to the requester (S104). This candidate list should preferably be sortable according to attributes or filterable based on more detailed conditions. As shown in FIG. 10, the candidate list (search results) is a list that compiles information on candidates extracted by the advertiser search system 100 as an analysis result, and includes items such as candidate information, candidate selection evaluation, and request / contact status. The candidate information includes the user name, profile information (gender, age group, etc.), number of followers, male / female ratio, engagement index, and attribute information (related categories, area, etc.). The candidate selection evaluation includes a suitability score and priority (ranking result) calculated by the system. The request / contact status includes whether or not a direct mail has been sent and the candidate's reply status. FIG. 4 is an image diagram showing an example of a database showing the list of candidates.

[0082] Next, the requester selects an appropriate candidate from the displayed candidates and sends a direct mail to the candidate via the DM sending unit 116.

[0083] Through this series of operations, the client can easily select candidates suitable for advertising their products, etc.

[0084] As shown in FIG. 12, the detailed processing flow of the advertiser search system of the present invention will be explained below step by step.

[0085] First, the client sets search conditions to identify appropriate advertisers needed to advertise their products or services (S201). These search conditions include, for example, the advertiser's attribute information, number of followers, engagement rate, and posting history related to specific keywords, and can be set arbitrarily by the client.

[0086] Next, based on the set conditions, the SNS information acquisition unit acquires various information about the advertiser candidates (S202). This acquired information includes profile information of the SNS accounts owned by each candidate and information posted using those accounts. In addition, depending on the embodiment, the email acquisition unit may acquire direct mail information sent and received by the candidate on the SNS, or may also acquire information about the client's own products or services, customer information of other parties competing with the client, etc.

[0087] Next, the command sending unit sends a command to the proposal system using the machine learning model using the acquired various information as input data (S203). This command can instruct the system to take into consideration, individually or in combination, client information (such as product / service features and marketing goals), candidate's SNS account information and posting information, PR rate, average number of likes, and number of engagements as indicators of the candidate's influence on SNS, as well as direct mail content, store information, advertiser attribute information, and competitor customer information.

[0088] Based on the above instructions, the machine learning model analyzes various information and evaluates and proposes advertiser candidates who are likely to meet the conditions specified by the client (S204). The evaluation results are generated in the form of a score indicating the degree of suitability, a priority order, etc.

[0089] Next, the information search unit receives the evaluation results output by the proposed system and extracts and narrows down the SNS account information that meets the search conditions set by the requester (S205). Specifically, it is also possible to perform filtering processing based on the attribute information contained in each SNS account information, making it possible to extract a limited number of candidates that match the conditions specified by the requester.

[0090] Furthermore, in some embodiments, the extracted advertiser candidates can be prioritized in order of advertising effectiveness based on the suitability calculated by the machine learning model (S206).

[0091] Finally, the direct mail sending unit sends direct mail to the advertiser candidates for whom the search and ranking have been completed, informing them of a request for cooperation in advertising activities, and direct contact can be made to carry out the advertising (S207).

[0092] Through the above-described flow, the advertiser search system of the present invention can efficiently and accurately search for and select the advertiser that best suits the advertising needs set by the client, thereby maximizing the advertising effect.

[0093] We will also explain the scraping process in the SNS information acquisition unit.

[0094] 13, the SNS information acquisition unit 108 of the advertiser search system 100 in this embodiment accesses the SNS server 400 and acquires the SNS information of the candidate by scraping. The detailed processing flow will be explained below step by step.

[0095] First, the SNS information acquisition unit 108 determines candidates to be scraped based on the conditions set by the client (S301). These conditions include attribute information such as the candidate's area of ​​activity, related categories, number of followers, or engagement rate.

[0096] Next, the SNS information acquisition unit 108 starts accessing the SNS server 400 to acquire the SNS information of the candidate determined as the target (S302). In this access process, an HTTP request is sent via a publicly available web page provided by the SNS server 400 or a predetermined API (Application Programming Interface) to acquire the target data.

[0097] Next, the SNS information acquisition unit 108 acquires the SNS account information of the target candidate (S303). The SNS account information acquired at this time includes the candidate's username (real name or handle name, etc.), gender, age group, related area (region in which the candidate is mainly active), related category (beauty, food and drink, fashion, etc.), number of followers, follower attributes, and various indices that can be used to evaluate the candidate's influence, such as PR rate, average number of likes, and number of engagements.

[0098] Furthermore, the SNS information acquisition unit 108 acquires the SNS posting information posted by the target candidate (S304). The acquired SNS posting information includes media data such as posted text articles, images, videos, and audio, as well as metadata including the posting date and time, posting location information (Geotag), hashtags used, filters, effect information, and the posting's visibility setting (for example, public, restricted access, etc.).

[0099] Furthermore, if necessary, the SNS information acquisition unit 108 can also acquire information on direct mail sent or received by the candidate on the SNS (S305). In this case, detailed email information is acquired, including the date and time of sending or receiving the direct mail, sender and recipient information, the email body, the presence or absence and content of attachments, the content of advertising requests received in the past, the presented remuneration conditions, whether or not a reply was made and the reply time, etc.

[0100] The acquired various information is then formatted into a predetermined format (S306) for efficient analysis and processing within the advertiser search system 100. The formatting process is performed by structuring the information in, for example, JSON format or XML format. The formatted data is then converted into a format that can be displayed on the user interface (UI) provided by the advertiser search system 100.

[0101] Next, the formatted data is stored in a database provided inside the advertiser search system 100 (S307). Specifically, the acquired SNS account information is stored in an SNS account information database, the SNS posting information is stored in an SNS posting information database, and the direct mail information is stored in a direct mail information database after being classified.

[0102] The scraping process ends when all information about the targeted candidates has been acquired and stored (S308).

[0103] After the acquired information is stored in the database, the information search unit 106, the command transmission unit 114, etc. start to sequentially evaluate and select candidates (S309).

[0104] Through the above-described processing flow, the SNS information acquisition unit 108 of the present invention automatically collects and analyzes information on potential advertisers efficiently and accurately, and provides an information base for extracting the best candidates to advertise the client's products and services.

[0105] The present invention is not limited to the above-described embodiment, and each configuration can be modified as appropriate within the scope of the present invention. [Explanation of symbols]

[0106] 100: Advertiser Search System 106: Information Retrieval Department 108: SNS information acquisition department 110: Mail Acquisition Unit 112: Customer information acquisition department 114: Command transmission unit 116: DM sending unit

Claims

1. an SNS information acquisition unit that acquires SNS account information of a plurality of candidates who can be advertisers and SNS post information posted using the SNS account information; an information search unit that searches for SNS account information that satisfies a condition designated by a client from the SNS account information and the SNS post information acquired by the SNS information acquisition unit; an email acquisition unit that acquires email information regarding the contents of direct mail from multiple candidates who can be advertisers; a command sending unit that sends a command to the proposal system using a machine learning model to search for an appropriate advertiser taking into account client information related to the client's product or service and the content of the email information acquired by the email acquisition unit; the information search unit searches for SNS account information that satisfies a condition designated by a requester from a result received in response to the command transmitted from the command transmission unit; The command sending unit analyzes information including store information and advertiser information in addition to the content of the email information acquired by the email acquisition unit, and sends a command to rank a plurality of candidates who can become advertisers. Advertiser search system.

2. an SNS information acquisition unit that acquires SNS account information of a plurality of candidates who can be advertisers and SNS post information posted using the SNS account information; an information search unit that searches for SNS account information that satisfies a condition designated by a client from the SNS account information and the SNS post information acquired by the SNS information acquisition unit; a customer information acquisition unit that acquires customer information of other parties competing with the client based on client information relating to the client's product or service; a command sending unit that sends a command to the proposal system using a machine learning model to search for an appropriate advertiser taking into account the customer information acquired by the customer information acquisition unit and the SNS account information acquired by the SNS information acquisition unit; the information search unit searches for SNS account information that satisfies a condition designated by a requester from a result received in response to the command transmitted from the command transmission unit. Advertiser search system.

3. A command sending unit sends a command to the proposal system using a machine learning model to search for an appropriate advertiser taking into account client information related to the client's product or service and the SNS account information acquired by the SNS information acquisition unit, the information search unit searches for SNS account information that satisfies a condition designated by a requester from a result received in response to the command transmitted from the command transmission unit.

3. The advertiser search system according to claim 1 or 2.

4. A command sending unit sends a command to the proposal system using a machine learning model to search for an appropriate advertiser taking into account client information related to the client's product or service and the SNS post information acquired by the SNS information acquisition unit, the information search unit searches for SNS account information that satisfies a condition designated by a requester from a result received in response to the command transmitted from the command transmission unit.

3. The advertiser search system according to claim 1 or 2.

5. The command sending unit sends a command to search for an appropriate advertiser taking into consideration at least one of a PR rate, an average number of likes, and an engagement number of a plurality of candidates who can become advertisers. The advertiser search system according to claim 4.

6. a direct mail sending unit that sends direct mail to a person associated with the searched SNS account information; 3. The advertiser search system according to claim 1 or 2.

7. The information search unit has a function of filtering search results based on attribute information included in the searched SNS account information.

3. The advertiser search system according to claim 1 or 2.

8. A computer-implemented advertiser search method, comprising: an SNS information acquisition step of acquiring SNS account information of a plurality of candidates who can become advertisers and SNS post information posted using the SNS account information; an information retrieval step of retrieving SNS account information that satisfies a condition designated by a requester from the SNS account information and the SNS posted information acquired in the SNS information acquisition step; an email acquisition step of acquiring email information regarding the contents of direct mail from a plurality of candidates who can be advertisers; a command sending step of sending a command to a proposal system using a machine learning model to search for an appropriate advertiser taking into consideration client information related to the client's product or service and the content of the email information acquired by the email acquisition unit; In the information search step, SNS account information that satisfies a condition designated by a requester is searched for from a result received in response to the command transmitted from the command transmission unit; In the command sending step, a command is sent to analyze information including store information and advertiser information in addition to the content of the email information acquired by the email acquisition unit, and to rank a plurality of candidates who can become advertisers. How to search for advertisers.

9. A computer-implemented advertiser search method, comprising: an SNS information acquisition step of acquiring SNS account information of a plurality of candidates who can become advertisers and SNS post information posted using the SNS account information; an information retrieval step of retrieving SNS account information that satisfies a condition designated by a requester from the SNS account information and the SNS posted information acquired in the SNS information acquisition step; a customer information acquisition step of acquiring customer information of other parties competing with the client based on client information relating to the client's product or service; a command sending step of sending a command to a proposal system using a machine learning model to search for an appropriate advertiser taking into account the customer information acquired in the customer information acquisition step and the SNS account information acquired in the SNS information acquisition step; In the information search step, SNS account information that satisfies a condition designated by a requester is searched for from a result received in response to the command transmitted from the command transmission unit. How to search for advertisers.

10. A program, A program for causing a computer to execute the advertiser search method according to claim 8 or 9.

Images