Analysis apparatus, analysis program, and analysis method

Abstract

To analyze the scale of information diffusion.SOLUTION: An analysis apparatus includes: a post information acquisition unit which acquires post information including information on normal posts including a predetermined keyword and diffusing posts that diffuses the normal posts, and stores the information as post groups in association with the diffusing posts related to the normal posts, the post groups including a small-scale diffusing post group in which the number of diffusing posts is smaller than a predetermined quantity, and a large-scale diffusing post group in which the number of diffusing posts is larger than the predetermined quantity; and an analysis unit which analyzes diffusing trends of the normal posts on the basis of information on the number of diffusing posts of the normal posts included in each of the small-scale diffusing post group and the large-scale diffusing post group.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an analysis apparatus, an analysis program, and an analysis method.

Background Art

[0002] The information diffusion phenomenon on the Internet is widely recognized as something that improves the recognition of products and commercial materials, and is also applied to practical effect measurement in digital marketing such as advertising distribution. However, the application of the information diffusion phenomenon is only recognized as a result, and it is difficult to predict such a phenomenon in advance. As long as there is such unpredictability, there is mainly a large cost inefficiency in the advertising production site, which has hindered the development of the industry.

[0003]

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] For example, Patent Document 1 discloses a technique for predicting the influence of each poster and the number of future posts on a specific topic on a website such as an SNS. However, the invention described in Patent Document 1 predicts the scale of the diffusion of primary information on the SNS for a specific topic, and it is difficult to predict the number of re-diffusions such as quoting a certain post and spreading it.

[0006]

[0007] ​​​​​​​The present invention has been made in view of such a background, and aims to analyze the scale of information diffusion. and so on.

Means for Solving the Problems

[0008] To solve the above problems, an analysis device according to the present disclosure acquires posting information including information on normal postings including a predetermined keyword and diffusion postings that diffuse each of the normal postings, and stores the diffusion postings related to the normal postings as a posting group in a posting information acquisition unit, the posting group includes a small-scale diffusion posting group in which the number of the diffusion postings is less than a predetermined number and a large-scale diffusion posting group in which the number of the diffusion postings is more than the predetermined number, and an analysis unit that analyzes the diffusion tendency of the normal postings based on information on the number of the diffusion postings of the normal postings included in each of the small-scale diffusion posting group and the large-scale diffusion posting group. and stores the diffusion postings related to the normal postings as a posting group in a posting information acquisition unit, the posting group includes a small-scale diffusion posting group in which the number of the diffusion postings is less than a predetermined number and a large-scale diffusion posting group in which the number of the diffusion postings is more than the predetermined number, and an analysis unit that analyzes the diffusion tendency of the normal postings based on information on the number of the diffusion postings of the normal postings included in each of the small-scale diffusion posting group and the large-scale diffusion posting group. and stores the diffusion postings related to the normal postings as a posting group in a posting information acquisition unit, the posting group includes a small-scale diffusion posting group in which the number of the diffusion postings is less than a predetermined number and a large-scale diffusion posting group in which the number of the diffusion postings is more than the predetermined number, and an analysis unit that analyzes the diffusion tendency of the normal postings based on information on the number of the diffusion postings of the normal postings included in each of the small-scale diffusion posting group and the large-scale diffusion posting group. and stores the diffusion postings related to the normal postings as a posting group in a posting information acquisition unit, the posting group includes a small-scale diffusion posting group in which the number of the diffusion postings is less than a predetermined number and a large-scale diffusion posting group in which the number of the diffusion postings is more than the predetermined number, and an analysis unit that analyzes the diffusion tendency of the normal postings based on information on the number of the diffusion postings of the normal postings included in each of the small-scale diffusion posting group and the large-scale diffusion posting group. and stores the diffusion postings related to the normal postings as a posting group in a posting information acquisition unit, the posting group includes a small-scale diffusion posting group in which the number of the diffusion postings is less than a predetermined number and a large-scale diffusion posting group in which the number of the diffusion postings is more than the predetermined number, and an analysis unit that analyzes the diffusion tendency of the normal postings based on information on the number of the diffusion postings of the normal postings included in each of the small-scale diffusion posting group and the large-scale diffusion posting group. and stores the diffusion postings related to the normal postings as a posting group in a posting information acquisition unit, the posting group includes a small-scale diffusion posting group in which the number of the diffusion postings is less than a predetermined number and a large-scale diffusion posting group in which the number of the diffusion postings is more than the predetermined number, and an analysis unit that analyzes the diffusion tendency of the normal postings based on information on the number of the diffusion postings of the normal postings included in each of the small-scale diffusion posting group and the large-scale diffusion posting group. and stores the diffusion postings related to the normal postings as a posting group in a posting information acquisition unit, the posting group includes a small-scale diffusion posting group in which the number of the diffusion postings is less than a predetermined number and a large-scale diffusion posting group in which the number of the diffusion postings is more than the predetermined number, and an analysis unit that analyzes the diffusion tendency of the normal postings based on information on the number of the diffusion postings of the normal postings included in each of the small-scale diffusion posting group and the large-scale diffusion posting group.

[0009] Regarding other problems disclosed in the present application and their solutions, they will be more clearly described in the section of the embodiments of the invention and the drawings. and so on.

Effects of the Invention

[0010] According to the present invention, the scale of information diffusion can be analyzed.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Mode for Carrying Out the Invention

[0012] <Summary of the Invention> [Item 1] Obtain post information including a normal post including a predetermined keyword and a diffusion post that diffuses each of the normal posts, associate the diffusion posts related to the normal posts, and use them as a post group and store it in a post information acquisition unit, The post group includes a small-scale diffusion post group in which the number of the diffusion posts is less than a predetermined number, and a large-scale diffusion post group in which the number of the diffusion posts is more than the predetermined number. Based on the information on the number of the diffusion posts of the normal posts included in each of the small-scale diffusion post group and the large-scale diffusion post group, an analysis unit that analyzes the diffusion tendency of the normal posts, An analysis device comprising: [Item 2] The analysis unit extracts so that the number of the diffusion posts in each of the small-scale diffusion post group and the large-scale diffusion post group becomes the same scale, and compares the diffusion posts that increase or decrease when the small-scale diffusion post group and the large-scale diffusion post group are increased or decreased by the same multiple, thereby analyzing the diffusion tendency. ​ The analyzer according to item 1. [Item 3] The post information acquisition unit acquires information on the accounts that made the normal post and the diffusion post, and evaluates the contribution degree of information diffusion of a specific account in the post group including the account that made the normal post and the account that made the diffusion post related to the normal post. An evaluation unit, and The analyzer according to item 1 or 2, comprising the above. [Item 4] A similarity determination unit that determines the similarity between the first keyword included in the normal post analyzed for the diffusion tendency and the second keyword for which the diffusion tendency is to be analyzed, based on the information on the number of diffusion posts; and The analyzer according to item 1 or 2, comprising the above. The analysis unit predicts the diffusion tendency of the second keyword based on the similarity. The analyzer according to item 1 or 2. [Item 5] A prediction unit that predicts the number of diffusions when the normal post is posted by the user of the account, based on the information on the diffusion tendency and the information on the contribution degree; and The analyzer according to item 3, comprising the above. [Item 6] A post information acquisition step of acquiring post information including a normal post containing a predetermined keyword and a diffusion post that diffuses each of the normal posts, associating the diffusion posts related to the normal post, and storing them as a post group; The post group includes a small-scale diffusion post group in which the number of diffusion posts is less than a predetermined number and a large-scale diffusion post group in which the number of diffusion posts is more than the predetermined number. Based on the information on the number of diffusion posts of the normal posts included in each of the small-scale diffusion post group and the large-scale diffusion post group, and ​​​​​An analysis step for analyzing the diffusion tendency of normal submissions, and An analysis program that causes the above to be executed. [Item 7] The computer Obtains submission information including normal submissions containing a predetermined keyword and diffusion submissions that diffuse each of the normal submissions, and links the diffusion submissions related to the normal submissions to form a submission group And stores it as a submission information acquisition step, The submission group includes a small-scale diffusion submission group in which the number of the diffusion submissions is less than a predetermined number and a large-scale diffusion submission group in which the number of the diffusion submissions is more than a predetermined number. Based on the information on the number of the diffusion submissions of the normal submissions included in each of the small-scale diffusion submission group and the large-scale diffusion submission group, An analysis step for analyzing the diffusion tendency of the normal submissions, and An analysis method that executes the above. An analysis step for analyzing the diffusion tendency of the normal submissions, and An analysis method that executes the above.

[0013] FIG. 1 is a diagram showing an example of the overall configuration of an evaluation system according to an embodiment of the present invention. The evaluation system of this embodiment Is configured to include a server device 1. The server device 1 is communicably connected to a user terminal 3 via a communication network 2. The communication network 2 is, for example, The Internet, and is constructed by a public telephone line network, a mobile phone line network, a wireless communication path, Ethernet (registered trademark), or the like.

[0014] ==Server device 1== The server device 1 may be a general-purpose computer such as a workstation or a personal computer, or may be logically implemented by cloud computing. In this embodiment, one is illustrated for convenience of explanation, but it is not limited to this, and a plurality of units may be used. It is not limited thereto, and a plurality of units may be used.

[0015] ==User terminal 3== The user terminal 3 is a computer used by a user who performs analysis. For example, a smart phone, a tablet computer, a personal computer, a wearable computer, etc. are examples. The user can access the server device 1 through, for example, an application executed on the user terminal 3 or a web browser.

[0016] Figure 2 is a diagram showing an example of the hardware configuration of the server device 1. Note that the illustrated configuration is an example, and it may have other configurations. The server device 1 includes a processor 101, a memory 102, a storage device 103, a communication interface 104, an input device 105, and an output device 106. The storage device 103 stores various data and programs, such as a hard disk drive, a solid state drive, a flash memory, etc. The communication interface 104 is an interface for connecting to the communication network 2, such as an adapter for connecting to Ethernet (registered trademark), a modem for connecting to a public telephone network, a wireless communication device for performing wireless communication, a USB (Universal Serial Bus) connector or an RS232C connector for serial communication, etc. The input device 105 inputs data, such as a keyboard, a mouse, a touch panel, a button, a microphone, etc. The output device 106 outputs data, such as a display, a printer, a speaker, etc. Note that each functional unit of the server device 1 described later is realized by the processor 101 reading a program stored in the storage device 103 into the memory 102 and executing it, and each storage unit of the server device 1 is realized as a part of the storage area provided by the memory 102 and the storage device 103. ​​​​​​​​​

[0017] Figure 3 shows the functional configuration of the server device 1. As shown in Figure 3, the server device 1 includes each storage unit of a contribution information storage unit 131 and a similar information storage unit 132, and a contribution information acquisition unit 1 11, an analysis unit 112, a contribution degree evaluation unit 113, a similarity determination unit 114, and a prediction unit 115 and a presentation unit 116.

[0018] Descriptions of each storage unit of the contribution information storage unit 131, the similar information storage unit 132, and the performance information storage unit 133 will be described.

[0019] The contribution information storage unit 131 stores information on contributions on a website on the Internet, an example of which is shown in Figure 4. The contribution information includes, as an example, contributions, re-contributions of contributions, quoted contributions (diffused contributions) that quote the contributions, the accounts that made the contributions and the diffused contributions, the followings of the accounts, the information of the followers, the types or numbers of evaluations for the contributions, etc., but is not limited thereto. Further, the contribution information includes a large-scale diffused contribution group and a small-scale diffused contribution group.

[0020] The similar information storage unit 132 stores the similar information of keywords or accounts determined by the similarity determination unit 114, an example of which is shown in Figure 5. The similar information storage unit 132 may store information on the degree of similarity in addition to the information on the similarity of keywords or accounts.

[0021] The performance information storage unit 133 stores the similar information of keywords or accounts determined by the similarity determination unit 114, an example of which is shown in Figure 6.

[0022] Hereinafter, the contribution information acquisition unit 111, the analysis unit 112, the contribution degree evaluation unit 113, and the similarity determination A description of the processing units including the unit 114, the prediction unit 115, and the presentation unit 116 will be given.

[0023] As an example, the posting information acquisition unit 111 acquires posting information from a website on the Internet via the communication network 2 and stores it in the posting information storage unit 131. The website from which the posting information acquisition unit 111 acquires posting information may, for example, have a function that allows a user to post information (hereinafter referred to as normal posting) from an account managed by the user, and an account managed by another user can re-post, quote-post (collectively referred to as diffusion posting such as re-posting and quote-posting), etc. It may be any social media such as a blog, X (former Twitter (registered trademark)), Instagram (registered trademark), TikTok (registered trademark), YouTube (registered trademark), 17Live (registered trademark), etc., but is not limited thereto. Note that the diffusion posting may include a case where the posting is displayed on the wall of the other user when the account managed by the other user indicates an evaluation of the normal posting.

[0024] The website from which the posting information acquisition unit 111 acquires posting information may, for example, have a function that allows a user to post information (hereinafter referred to as normal posting) from an account managed by the user, and an account managed by another user can re-post, quote-post (collectively referred to as diffusion posting such as re-posting and quote-posting), etc. It may be any social media such as a blog, X (former Twitter (registered trademark)), Instagram (registered trademark), TikTok (registered trademark), YouTube (registered trademark), 17Live (registered trademark), etc., but is not limited thereto. Note that the diffusion posting may include a case where the posting is displayed on the wall of the other user when the account managed by the other user indicates an evaluation of the normal posting. The website from which the posting information acquisition unit 111 acquires posting information may, for example, have a function that allows a user to post information (hereinafter referred to as normal posting) from an account managed by the user, and an account managed by another user can re-post, quote-post (collectively referred to as diffusion posting such as re-posting and quote-posting), etc. It may be any social media such as a blog, X (former Twitter (registered trademark)), Instagram (registered trademark), TikTok (registered trademark), YouTube (registered trademark), 17Live (registered trademark), etc., but is not limited thereto. Note that the diffusion posting may include a case where the posting is displayed on the wall of the other user when the account managed by the other user indicates an evaluation of the normal posting. The website from which the posting information acquisition unit 111 acquires posting information may, for example, have a function that allows a user to post information (hereinafter referred to as normal posting) from an account managed by the user, and an account managed by another user can re-post, quote-post (collectively referred to as diffusion posting such as re-posting and quote-posting), etc. It may be any social media such as a blog, X (former Twitter (registered trademark)), Instagram (registered trademark), TikTok (registered trademark), YouTube (registered trademark), 17Live (registered trademark), etc., but is not limited thereto. Note that the diffusion posting may include a case where the posting is displayed on the wall of the other user when the account managed by the other user indicates an evaluation of the normal posting. The website from which the posting information acquisition unit 111 acquires posting information may, for example, have a function that allows a user to post information (hereinafter referred to as normal posting) from an account managed by the user, and an account managed by another user can re-post, quote-post (collectively referred to as diffusion posting such as re-posting and quote-posting), etc. It may be any social media such as a blog, X (former Twitter (registered trademark)), Instagram (registered trademark), TikTok (registered trademark), YouTube (registered trademark), 17Live (registered trademark), etc., but is not limited thereto. Note that the diffusion posting may include a case where the posting is displayed on the wall of the other user when the account managed by the other user indicates an evaluation of the normal posting. The website from which the posting information acquisition unit 111 acquires posting information may, for example, have a function that allows a user to post information (hereinafter referred to as normal posting) from an account managed by the user, and an account managed by another user can re-post, quote-post (collectively referred to as diffusion posting such as re-posting and quote-posting), etc. It may be any social media such as a blog, X (former Twitter (registered trademark)), Instagram (registered trademark), TikTok (registered trademark), YouTube (registered trademark), 17Live (registered trademark), etc., but is not limited thereto. Note that the diffusion posting may include a case where the posting is displayed on the wall of the other user when the account managed by the other user indicates an evaluation of the normal posting. The website from which the posting information acquisition unit 111 acquires posting information may, for example, have a function that allows a user to post information (hereinafter referred to as normal posting) from an account managed by the user, and an account managed by another user can re-post, quote-post (collectively referred to as diffusion posting such as re-posting and quote-posting), etc. It may be any social media such as a blog, X (former Twitter (registered trademark)), Instagram (registered trademark), TikTok (registered trademark), YouTube (registered trademark), 17Live (registered trademark), etc., but is not limited thereto. Note that the diffusion posting may include a case where the posting is displayed on the wall of the other user when the account managed by the other user indicates an evaluation of the normal posting. The website from which the posting information acquisition unit 111 acquires posting information may, for example, have a function that allows a user to post information (hereinafter referred to as normal posting) from an account managed by the user, and an account managed by another user can re-post, quote-post (collectively referred to as diffusion posting such as re-posting and quote-posting), etc. It may be any social media such as a blog, X (former Twitter (registered trademark)), Instagram (registered trademark), TikTok (registered trademark), YouTube (registered trademark), 17Live (registered trademark), etc., but is not limited thereto. Note that the diffusion posting may include a case where the posting is displayed on the wall of the other user when the account managed by the other user indicates an evaluation of the normal posting. The website from which the posting information acquisition unit 111 acquires posting information may, for example, have a function that allows a user to post information (hereinafter referred to as normal posting) from an account managed by the user, and an account managed by another user can re-post, quote-post (collectively referred to as diffusion posting such as re-posting and quote-posting), etc. It may be any social media such as a blog, X (former Twitter (registered trademark)), Instagram (registered trademark), TikTok (registered trademark), YouTube (registered trademark), 17Live (registered trademark), etc., but is not limited thereto. Note that the diffusion posting may include a case where the posting is displayed on the wall of the other user when the account managed by the other user indicates an evaluation of the normal posting. The website from which the posting information acquisition unit 111 acquires posting information may, for example, have a function that allows a user to post information (hereinafter referred to as normal posting) from an account managed by the user, and an account managed by another user can re-post, quote-post (collectively referred to as diffusion posting such as re-posting and quote-posting), etc. It may be any social media such as a blog, X (former Twitter (registered trademark)), Instagram (registered trademark), TikTok (registered trademark), YouTube (registered trademark), 17Live (registered trademark), etc., but is not limited thereto. Note that the diffusion posting may include a case where the posting is displayed on the wall of the other user when the account managed by the other user indicates an evaluation of the normal posting.

[0025] As an example, the posting information acquisition unit 111 acquires posting information including a predetermined keyword. The keyword may include, but is not limited to, a product name, a service, a talent (including but not limited to entertainers, youtubers, v-tubers, streamers, etc.), the official name or abbreviation or nickname of a character name, a generalized commercial product name (e.g., web service, confectionery, beverage, food, cosmetics, real estate, etc.), an activity, an event, an industry name (sports, fashion, dining out, etc.), etc. The keyword may include, but is not limited to, a product name, a service, a talent (including but not limited to entertainers, youtubers, v-tubers, streamers, etc.), the official name or abbreviation or nickname of a character name, a generalized commercial product name (e.g., web service, confectionery, beverage, food, cosmetics, real estate, etc.), an activity, an event, an industry name (sports, fashion, dining out, etc.), etc. The keyword may include, but is not limited to, a product name, a service, a talent (including but not limited to entertainers, youtubers, v-tubers, streamers, etc.), the official name or abbreviation or nickname of a character name, a generalized commercial product name (e.g., web service, confectionery, beverage, food, cosmetics, real estate, etc.), an activity, an event, an industry name (sports, fashion, dining out, etc.), etc. The keyword may include, but is not limited to, a product name, a service, a talent (including but not limited to entertainers, youtubers, v-tubers, streamers, etc.), the official name or abbreviation or nickname of a character name, a generalized commercial product name (e.g., web service, confectionery, beverage, food, cosmetics, real estate, etc.), an activity, an event, an industry name (sports, fashion, dining out, etc.), etc. The keyword may include, but is not limited to, a product name, a service, a talent (including but not limited to entertainers, youtubers, v-tubers, streamers, etc.), the official name or abbreviation or nickname of a character name, a generalized commercial product name (e.g., web service, confectionery, beverage, food, cosmetics, real estate, etc.), an activity, an event, an industry name (sports, fashion, dining out, etc.), etc.

[0026] Note that the posting information acquired by the posting information acquisition unit 111 includes UGC (user generated content). UGC is a word indicating a reference to a product or the like by a certain consumer. Starting from the UGC posted from a certain account, cases are known where followers and others make diffused posts, thereby improving the recognition of the product or the like.

[0027] The posting information acquisition unit 111 may acquire the above-described posting information by specifying the language in which the post was made and the period to be extracted.

[0028] The posting information acquisition unit 111 may acquire the accounts that made normal posts and diffused posts, and the follow and follower information of those accounts.

[0029] The posting information acquisition unit 111 extracts normal posts and diffused posts made based on the normal posts, and generates a posting group.

[0030] The posting information acquisition unit 111 classifies a posting group with the number of diffused posts less than a predetermined number as a small-scale diffused posting group, and a posting group with the number of diffused posts more than a predetermined number as a large-scale diffused posting group, and stores them in

[0031] the posting information storage unit 131. Based on the follow / follower relationship between the account that made a normal post and the account that diffused the normal post, the posting information acquisition unit 111 extracts the network (posting group) of accounts regarding the diffusion of the normal post. In this case, the posting information acquisition unit

[0032] 111 may extract the posting group using a known method such as the Leiden method, for example.

[0032] The posting information acquisition unit 111 determines the size of the posting group, that is, the number of Classify the posting groups. For example, the size of the posting group can be classified according to the number of accounts included in the posting group, such as 1 (m icro), 2 - 10 (small), 11 - 99 (medium), 100 or more (la rge), etc. The posting information acquisition unit 111 can further classify the posting groups with the number of accounts included in the posting group being, for example, 1, 2 - 10 as small-scale diffusion posting groups, 11 - 99, 100 or more as large-scale diffusion posting groups, but is not limited to these thresholds. If it is assumed that 1 account has made a diffusion post once, the number of accounts mentioned here can be regarded as the number of posts. Therefore, the number of accounts that have made diffusion posts can be regarded as the number of diffusion posts. As an example, the analysis unit 112 analyzes the diffusion tendency of normal posts based on the information on the total number of posts including the diffusion posts of normal posts included in each of the small-scale diffusion posting group and the large-scale diffusion posting group. As an example, based on the information on the number of diffusion posts included in each of the small-scale diffusion posting group and the large-scale diffusion posting group for normal posts containing a certain keyword (referred to as keyword A), the analysis unit 112 calculates the average value or median value of the diffusion posts of each of the small-scale diffusion posting group and the large-scale diffusion posting group, and stores it in the performance information storage unit 133. As shown in FIG. 7, for example, the analysis unit 112 extracts the small-scale diffusion posting group and the large-scale diffusion posting group such that the total number of posts of each of them including normal posts containing keyword A is of the same scale. The analysis unit 112 samples and extracts a part of the normal posts of each of these small-scale diffusion posting groups and large-scale diffusion posting groups. The analysis unit 112 is for the small-scale diffusion posting group and the large-scale diffusion posting

[0033] As an example, the analysis unit 112 analyzes the diffusion tendency of normal posts based on the information on the total number of posts including the diffusion posts of normal posts included in each of the small-scale diffusion posting group and the large-scale diffusion posting group. As an example, based on the information on the number of diffusion posts included in each of the small-scale diffusion posting group and the large-scale diffusion posting group for normal posts containing a certain keyword (referred to as keyword A), the analysis unit 112 calculates the average value or median value of the diffusion posts of each of the small-scale diffusion posting group and the large-scale diffusion posting group, and stores it in the performance information storage unit 133. .

[0034] As an example, the analysis unit 112 is based on the information on the number of diffusion posts included in each of the small-scale diffusion posting group and the large-scale diffusion posting group for normal posts containing a certain keyword (referred to as keyword A). Based on the information on the number of diffusion posts included in each of the small-scale diffusion posting group and the large-scale diffusion posting group for normal posts containing a certain keyword (referred to as keyword A), the analysis unit 112 calculates the average value or median value of the diffusion posts of each of the small-scale diffusion posting group and the large-scale diffusion posting group, and stores it in the performance information storage unit 133. Based on the information on the number of diffusion posts included in each of the small-scale diffusion posting group and the large-scale diffusion posting group for normal posts containing a certain keyword (referred to as keyword A), the analysis unit 112 calculates the average value or median value of the diffusion posts of each of the small-scale diffusion posting group and the large-scale diffusion posting group, and stores it in the performance information storage unit 133. Calculate the average value or median value of the diffusion posts of each of the small-scale diffusion posting group and the large-scale diffusion posting group, and store it in the performance information storage unit 133.

[0035] As shown in FIG. 7, for example, the analysis unit 112 extracts the small-scale diffusion posting group and the large-scale diffusion posting group such that the total number of posts of each of them including normal posts containing keyword A is of the same scale. The analysis unit 112 samples and extracts a part of the normal posts of each of these small-scale diffusion posting groups and large-scale diffusion posting groups. The analysis unit 112 samples and extracts a part of the normal posts of each of these small-scale diffusion posting groups and large-scale diffusion posting groups. The analysis unit 112 samples and extracts a part of the normal posts of each of these small-scale diffusion posting groups and large-scale diffusion posting groups. When the normal posts sampled from the group are increased or decreased by the same multiple, the value of the average or median of the diffusion posts of the normal posts containing the above-mentioned keyword A is read from the performance information storage unit 133 and multiplied by the number of normal posts, and by comparing the number of the total number of posts to be increased or decreased, the diffusion tendency of the normal posts containing keyword A is analyzed.

[0036] For example, for a keyword (referred to as keyword B) for which the analysis unit 112 has not acquired data on the average or median of the diffusion posts, the analysis unit 112 extracts the keyword so that the total number of posts in each of the small-scale diffusion post group and the large-scale diffusion post group becomes approximately the same scale. The analysis unit 112 samples and extracts a part of the normal posts in each of these small-scale diffusion post groups and large-scale diffusion post groups. The analysis unit 112, when the normal posts sampled from the small-scale diffusion post group and the large-scale diffusion post group are increased or decreased by the same multiple, the value of the average or median of the diffusion posts of the normal posts containing keyword A that the similarity determination unit 114 described later determines to be similar to keyword B (assuming that the similarity determination unit 114 determines that keyword B is similar to keyword A) is read from the performance information storage unit 133, multiplied by the number of normal posts, and by comparing the number of the total number of posts to be increased or decreased, the diffusion tendency of the normal posts containing keyword B is analyzed.

[0037] A specific example of the analysis by the analysis unit 112 will be described. The analysis unit 112 extracts a small-scale diffusion post group (total number of posts 1000, number of post groups 100, average number of diffusion posts per normal post is 9) and a large-scale diffusion post group (total number of posts 900, number of post groups 20, average number of diffusion posts per normal post is 44) for the normal posts containing keyword A (the small-scale diffusion post group and the large-scale diffusion post group The total number of submissions for both is 1900). The analysis unit 112 extracts, for example, a small-scale diffusion submission group of 25 (total number of submissions is 250) and a large-scale diffusion submission group of 5 (total number of submissions is 225), so that the number of submissions in each of the small-scale diffusion submission group and the large-scale diffusion submission group is made comparable. The analysis unit 112 randomly extracts, for example, 20% (5 normal submissions in the small-scale diffusion submission group and 1 normal submission in the large-scale diffusion submission group) from the normal submissions of each of the extracted small-scale diffusion submission group and large-scale diffusion submission group. The analysis unit 112 multiplies the extracted normal submissions, for example, ( resulting in 30 normal submissions in the small-scale diffusion submission group and 6 normal submissions in the large-scale diffusion submission group), and also increases the number of diffusion submissions associated with the multiplied normal submissions at the same ratio. In this case, for the 5 increased normal submissions in the small-scale diffusion submission group, the number of diffusion submissions increases by 45, and the total number of submissions increases by 50. Also, the total number of submissions for 30 normal submissions is 300. For the 1 increased normal submission in the large-scale diffusion submission group, the number of diffusion submissions increases by 44, and the total number of submissions increases by 45. Also, the total number of submissions for 6 normal submissions is 270. Therefore, as the diffusion tendency of normal submissions containing keyword A, increasing the normal submissions in the small-scale diffusion submission group makes it easier to increase the total number of submissions. By performing the above-described processing, the analysis unit 112 analyzes which of the small-scale diffusion submission group or the large-scale diffusion submission group forms a larger total number of submissions when the diffusion submissions based on normal submissions containing a specific keyword are considered. Note that in the above example, the average number of diffusion submissions is the key stored in the performance information storage unit 133.

[0038] ​An example of a normal post containing the word A was introduced, but for the keyword B, regarding which the analysis unit 112 uses the information on the average number of diffusion posts of the keyword A determined by the similarity determination unit 114 to be similar to the keyword B, and calculates the increase in the number of diffusion posts based on the normal post when the normal post is increased. It is sufficient to calculate the increase in the number of diffusion posts based on the normal post.

[0039] Figure 8 is a diagram showing an example of the diffusion tendency analyzed by the analysis unit 112. The merchandise in Figure 8 includes specific product names or general merchandise names, and it can be analyzed whether the total number of posts increases more for the small-scale diffusion post group or the large-scale diffusion post group among the normal posts including these as keywords. For example, for snacks_1 and snacks_2, the diffusion by the small-scale post group shows that the total number of posts increases more, while for snacks _3 and snacks_4, the diffusion by the large-scale post group shows that the total number of posts increases more.

[0040] As an example, the analysis unit 112 may predict the diffusion tendency of a normal post including a certain keyword based on the information on the similarity of the keywords included in the normal post determined by the similarity determination unit 114 described later. The analysis unit 112 may analyze that keywords with high similarity have the same diffusion tendency.

[0041] As an example, the analysis unit 112 may predict the diffusion tendency of a normal post by a certain account based on the information on the similarity of the accounts determined by the similarity determination unit 114 described later. The analysis unit 112 may analyze that normal posts by accounts with high similarity have the same diffusion tendency.

[0042] As an example, the contribution degree evaluation unit 113, for the account that made a normal post and the ​​​​​​​​​Evaluate the contribution degree of information diffusion of a specific account in a group of posts including the account that made the relevant diffusion post. Evaluate the contribution degree.

[0043] For example, as shown in FIG. 9, the contribution degree evaluation unit 113 analyzes the post group and estimates an information diffusion network indicating how each normal post is diffusely posted. Specifically, the contribution degree evaluation unit 113 extracts all the accounts that diffusely posted a certain normal post and sets it as the node set V of the network. Further, when the contribution degree evaluation unit 113 sets the node set of the accounts that made normal posts or diffusion posts as Vfrom and the node set of the accounts that received normal posts or diffusion posts as Vto, it creates an edge between Vfrom and Vto and estimates the information diffusion network. Further, the contribution degree evaluation unit 113 analyzes how the number of diffusion posts changes as a result of excluding the account (evaluation account) for which the contribution degree to diffusion is to be evaluated from the post group. Specifically, the contribution degree evaluation unit 113 randomly extracts an evaluation account from the post group and excludes the node corresponding to the evaluation account from the diffusion path network. Next, the contribution degree evaluation unit 113 also excludes the nodes of the accounts to which information has diffused from the evaluation account and repeats this. Finally, the contribution degree evaluation unit 113 sets the number of excluded nodes as the contribution degree of the evaluation account to information diffusion. Estimate the information diffusion network showing how the post was diffusely posted. Specifically, The contribution degree evaluation unit 113 extracts all the accounts that diffusely posted a certain normal post and sets it as the node set V of the network. Further, when the contribution degree evaluation unit 113 sets the node set of the accounts that made normal posts or diffusion posts as Vfrom and the node set of the accounts that received normal posts or diffusion posts as Vto, it creates an edge between Vfrom and Vto and estimates the information diffusion network. Further, the contribution degree evaluation unit 113 analyzes how the number of diffusion posts changes as a result of excluding the account (evaluation account) for which the contribution degree to diffusion is to be evaluated from the post group. Specifically, the contribution degree evaluation unit 113 randomly extracts an evaluation account from the post group and excludes the node corresponding to the evaluation account from the diffusion path network. Next, the contribution degree evaluation unit 113 also excludes the nodes of the accounts to which information has diffused from the evaluation account and repeats this. Finally, the contribution degree evaluation unit 113 sets the number of excluded nodes as the contribution degree of the evaluation account to information diffusion. Exclude the node corresponding to the evaluation account from the diffusion path network. Next, the contribution degree evaluation unit 113 also excludes the nodes of the accounts to which information has diffused from the evaluation account and repeats this. Finally, the contribution degree evaluation unit 113 sets the number of excluded nodes as the contribution degree of the evaluation account to information diffusion.

[0044] As an example, the similarity determination unit 114 determines whether the first keyword included in the normal post analyzed for the diffusion tendency is similar to the second keyword for which the diffusion tendency is to be analyzed. Determine whether they are similar.

[0045] For example, the similarity determination unit 114 determines based on categories such as products and services in which the keyword is included, the key Based on the information of the characteristics of the products, services, etc. indicated by the keywords, the similarity between the keywords is determined. For example, when the first keyword is the product name of cosmetics and the second keyword is also the product name of cosmetics, the similarity determination unit 114 determines that the first keyword and the second keyword are similar. Also, for example, when the first keyword is the product name of snacks, the second keyword is also the product name of snacks, and each product has a sweet characteristic or contains chocolate, or has the same selling price, etc., when there are common characteristics the similarity determination unit 114 may determine that the first keyword and the second keyword are similar.

[0046] As an example, the similarity determination unit 114 may determine the similarity of accounts.

[0047] For example, the similarity determination unit 114 analyzes the graph structure representing the relationship of accounts, and may determine that accounts with similar graph structures are similar. Note that the graph structure is a graph structure represented by nodes (accounts) and edges (relationships connecting accounts), which includes the relationships between accounts in social media (for example, follow, follower relationships, relationships such as evaluation (likes, etc.) and diffusion to the posts of a certain account, and relationships associated with replies, etc.). Also, when determining the similarity of the graph structure, the overlap of edges in important node / edge relationships, the similarity degree of central node (account) attributes, the centrality value of nodes calculated by typical algorithms such as PageRank, etc. can be considered, but it is not limited to this. relationship, relationship such as evaluation (likes, etc.) and diffusion to the posts of a certain account, and relationship associated with replies, etc.), including), and the graph structure represented by nodes (accounts) and edges (relationships connecting accounts) is sufficient. In addition, when determining the similarity of the graph structure, the overlap of edges in important node / edge relationships, the similarity degree of central node (account) attributes, the centrality value of nodes calculated by typical algorithms such as PageRank, etc. can be considered, but it is not limited to this. In addition, when determining the similarity of the graph structure, the overlap of edges in important node / edge relationships, the similarity degree of central node (account) attributes, the centrality value of nodes calculated by typical algorithms such as PageRank, etc. can be considered, but it is not limited to this. degree of node (account) centrality calculated by typical algorithms such as PageRank, etc. can be considered, but it is not limited to this. The similarity determination unit 114 may consider, but is not limited to, the overlap of edges in important node / edge relationships, the similarity degree of central node (account) attributes, the centrality value of nodes calculated by typical algorithms such as PageRank, etc. when determining the similarity of the graph structure.

[0048] The similarity determination unit 114 may consider the situation of the entity (including individuals, enterprises, etc.) operating the account. Obtain reports and determine the similarity between accounts based on information such as business content, industry, business type, products handled, target industries, and business scale. Also, the similarity determination unit 114 may determine the similarity between accounts based on information such as the service sites or home pages operated by the business entity, comparing the design of the sites (keywords / phrases used, texture, color, font type, size, etc.) and determining the similarity between accounts based on the similarity of the design.

[0049] The prediction unit 115, for example, predicts how to increase diffusion posts when a normal post is made from a certain account based on information on diffusion tendency and contribution degree, or based on information on diffusion tendency, contribution degree, and further similarity degree.

[0050] The prediction unit 115, for example, predicts how to increase diffusion posts when a normal post is made from a certain account based on information on diffusion tendency and contribution degree. The prediction unit 115 obtains from the analysis result of the analysis unit 112 which of the small-scale diffusion post group and the large-scale diffusion post group is more likely to obtain the scale of diffusion posts for a normal post containing a certain keyword. Further, the prediction unit 115 may select a post group including accounts with a high contribution degree for the normal post.

[0051] The prediction unit 115, for example, predicts how to increase diffusion posts when a normal post is made from a certain account based on information on diffusion tendency, contribution degree, and similarity degree. The prediction unit 115 obtains information on the keyword for which it wants to predict how to increase diffusion posts by accepting user input operations. The prediction unit 115 uses the determination result of the similarity determination unit 114. ​​​​​​​​​​​​​​Obtain information on keywords with a high degree of similarity to the keyword. The prediction unit 115 is the analysis unit 1 From the results of the analysis in 12, for normal posts containing keywords with a high degree of similarity, the diffusion tendency is to obtain which of the small-scale diffusion post group and the large-scale diffusion post group is more likely to obtain the scale of diffusion posts . Furthermore, the prediction unit 115 may select a post group including accounts with a high contribution degree for normal posts and diffusion posts .

[0052] The presentation unit 116 presents to the user terminal 3 a method for increasing the number of diffusion posts in the case where a normal post is posted from a certain account predicted by the prediction unit 115 .

[0053] FIG. 10 is a diagram for explaining an example of the processing of the evaluation apparatus of the present embodiment

[0054] The server device 1 acquires post information (1001). The server device 1 analyzes the post information (1002). The server device 1 evaluates the contribution degree of the account to the diffusion post ( 1003). The server device 1 determines the similarity of the keyword or the account (100 4). The server device 1 predicts a method for obtaining more diffusion posts from normal posts (1005). The server device 1 presents the predicted method to the user terminal 3 (1006).

[0055] As described above, the preferred embodiments of the present disclosure 1 have been described in detail with reference to the accompanying drawings. However, the technical scope of the present disclosure is not limited to such examples. For those having ordinary knowledge in the technical field of the present disclosure , it is obvious that various variations or modification examples can be conceived within the scope of the technical idea described in the claims, and these are also naturally understood to belong to the technical scope of the present disclosure .

[0056] The devices described in this specification may be implemented as a single device, or may be implemented by a plurality of devices (such as cloud servers) partially or entirely connected by a communication network. For example, the CPU and storage device of server device 1 may be implemented by different servers connected to each other by a communication network. For example, the CPU and storage device of server device 1 may be implemented by different servers connected to each other by a communication network.

[0057] A series of processes performed by the devices described in this specification may be implemented using any of software, hardware, and a combination of software and hardware. A computer program for implementing each function of server device 1 according to this embodiment can be created and installed on a PC or the like. Also, a computer-readable recording medium on which such a computer program is stored can be provided. The recording medium is, for example, a magnetic disk, an optical disk, a magneto-optical disk, a flash memory, or the like. Further, the above-described computer program may be distributed via a communication network, for example, without using a recording medium. For example, a magnetic disk, an optical disk, a magneto-optical disk, a flash memory, or the like. Also, the above-described computer program may be distributed via a communication network, for example, without using a recording medium.

[0058] Also, the processes described in this specification do not necessarily have to be executed in the described order. Some process steps may be executed in parallel. Also, additional process steps may be adopted, and some process steps may be omitted.

[0059] Also, the effects described in this specification are merely illustrative or exemplary and are not limiting. That is, the technology according to the present disclosure may exhibit other effects apparent to those skilled in the art from the description of this specification, together with or instead of the above-described effects.

Description of Reference Numerals

[0060] 1 Server device 2 Communication network 3 User terminal 101 CPU 102 Memory 103 Storage device 104 Communication interface 105 Input device 106 Output device 111 Submission information acquisition unit 112 Analysis unit 113 Contribution degree evaluation unit 114 Similarity determination unit 115 Prediction unit 116 Presentation unit 131 Submission information storage unit 132 Similar information storage unit 133 Performance information storage unit

Claims

1. A normal post including a predetermined keyword and a spreading post that spreads each of the normal posts; The post information including the information is acquired, and the spread post related to the normal post is linked to the spread post to form a post group. A post information acquisition unit that stores the post information; The post group includes a small-scale spread post group having a number of the spread posts less than a predetermined number, and a spread post group having a number of the spread posts less than a predetermined number. A large-scale spread posting group in which the number of posts is greater than a predetermined number is included, and the small-scale spread posting group and the large-scale spread posting group are Based on the information on the number of diffusion posts of the normal post included in each of the groups of diffusion posts, An analysis department that analyzes the diffusion trends of regular posts; An analysis device comprising:

2. The analysis unit analyzes the diffusion contributions of the small-scale diffusion post group and the large-scale diffusion post group. The number of posts is extracted so that it is of the same scale, and the small-scale spread post group and the large-scale spread post group are extracted. By comparing the increase or decrease in the spread posts when the group is increased or decreased by the same factor, the spread tendency is Analyze, The analytical device of claim 1 .

3. The post information acquisition unit acquires information of the accounts that have posted the normal post and the diffusion post. Get The account that posted the normal post and the account that posted the diffusion post related to the normal post An evaluation to evaluate the contribution of a specific account to the spread of information in the group of posts including the The value part, The analysis device according to claim 1 or 2, comprising:

4. Based on the information on the number of spreading posts, the first The similarity between the first keyword and the second keyword whose diffusion tendency is to be analyzed is determined. A similarity determination unit Equipped with The analysis unit predicts the spreading tendency of the second keyword based on the similarity.

3. The analysis device according to claim 1 or 2.

5. Based on the information on the diffusion tendency and the information on the contribution degree, the normal post is A prediction unit that predicts the number of spreads when the user posts; The analysis device of claim 3 .

6. On the computer, A normal post including a predetermined keyword and a spreading post that spreads each of the normal posts; The post information including the information is acquired, and the spread post related to the normal post is linked to the spread post to form a post group. a post information acquisition step for storing the post information; The post group includes a small-scale spread post group having a number of the spread posts less than a predetermined number, and a spread post group having a number of the spread posts less than a predetermined number. A large-scale spread posting group in which the number of posts is greater than a predetermined number is included, and the small-scale spread posting group and the large-scale spread posting group are Based on the information on the number of diffusion posts of the normal post included in each of the groups of diffusion posts, An analysis step of analyzing the diffusion trend of the regular posts; An analysis program that runs

7. The computer A normal post including a predetermined keyword and a spreading post that spreads each of the normal posts; The post information including the information is acquired, and the spread post related to the normal post is linked to the spread post to form a post group. a post information acquisition step for storing the post information; The post group includes a small-scale spread post group having a number of the spread posts less than a predetermined number, and a spread post group having a number of the spread posts less than a predetermined number. A large-scale spread posting group having a number of posts greater than a predetermined number is included, and the small-scale spread posting group and the large-scale spread posting group are Based on the information on the number of diffusion posts of the normal post included in each of the groups of diffusion posts, An analysis step of analyzing the diffusion trend of the regular posts; ,Perform ,analysis ,methods.

Images