Social media information dissemination evaluation method and device

Abstract

The present application discloses a method and device for evaluating the dissemination of social media information. A specific implementation of the method includes: obtaining users related to the current social media information and the relationships between each user; calculating the influence score of each user based on the relationships between each user; obtaining the interactive content of each user related to the current social media information based on the relationships between each user; calculating the quality score of each user's interactive content based on the interactive content of each user; and calculating the dissemination evaluation score of the current social media information based on the influence score of each user and the quality score of each user's interactive content. This implementation can effectively measure the dissemination effect of social media information.

Description

Technical Field

[0001] The present disclosure relates to the field of Internet technology, and in particular to a social media information dissemination evaluation method, device, electronic device, and computer-readable medium. Background Art

[0002] Currently, there are billions of users in the world using social media, such as Facebook, Twitter, and Weibo. Many "big Vs (Weibo users with fans who have been certified by Weibo)" and "KOLs (Key Opinion Leaders)" generate a large number of topics every day through social media, influencing the minds of millions of users. However, social media information is different from traditional Internet advertising, and it is not appropriate to use traditional methods to evaluate its effectiveness. Summary of the invention

[0003] The embodiments of the present disclosure provide a method, an apparatus, an electronic device, and a computer-readable medium for evaluating the dissemination of social media information.

[0004] In a first aspect, an embodiment of the present disclosure provides a method for evaluating the dissemination of social media information, the method comprising: obtaining users related to current social media information and the relationships between each user; calculating the influence score of each user based on the users and the relationships between each user; obtaining the interactive content of each user related to the current social media information based on the relationships between each user; calculating the quality score of the interactive content of each user based on the interactive content of each user; and calculating the dissemination evaluation score of the current social media information based on the influence score of each user and the quality score of the interactive content of each user.

[0005] In some embodiments, each user of the above method includes: an interactive user related to the current social media information and an associated user related to the interactive user; the relationship between each user includes: the relationship between the interactive user and the current social media information, and the relationship between the interactive user and the associated user; the above calculation of the influence score of each user based on each user and the relationship between each user includes: establishing a user relationship network based on the interactive users, the associated users, the relationship between the interactive users and the current social media information, and the relationship between the interactive users and the associated users; and determining the influence score of each user in the user relationship network based on the user relationship network.

[0006] In some embodiments, the interactive content includes: interactive information and recognition; the above-mentioned calculation of the quality score of each user's interactive content based on the interactive content of each user includes: inputting the interactive information of each user into a trained sentiment analysis model to obtain a probability value of the interactive information of each user output by the sentiment analysis model being positive content; inputting the probability value and recognition of the positive content into a quality score formula to calculate the quality score of the interactive content of each user; the quality score formula is: S X,A =Min(1.0,P X,A -μ+L X,A / ε), where S X,A is the quality score of the interactive content of user X; L X,A is the user X’s recognition of the current social media information A advertisement, L X,A =0 means not approved, L X,A =1 means approval; P X,A ∈[0,1.0] represents the probability value that the interactive information of user X is positive content; μ and ε are specified coefficients.

[0007] In some embodiments, the above-mentioned interactive content includes: recognition; the above-mentioned calculation of the quality score of each user's interactive content based on the interactive content of each user includes: for each user, in response to determining that the user has recognition, determining that the quality score of the user's interactive content is: 0.5-μ+1 / ε, where μ and ε are specified coefficients.

[0008] In some embodiments, the above-mentioned calculation of the dissemination evaluation score of the current social media information based on the influence score of each user and the quality score of the interactive content of each user includes: multiplying the influence score of each user by the quality score of the corresponding interactive content to obtain the dissemination evaluation score of each user; adding the dissemination evaluation scores of all users to obtain the dissemination evaluation score of the current social media information.

[0009] In some embodiments, it also includes: determining core users among users based on the influence scores of each user and the quality scores of the interactive content of each user.

[0010] In some embodiments, the above-mentioned determination of core users among users based on the influence score of each user and the quality score of the interactive content of each user includes: multiplying the influence score of each user by the quality score of the corresponding interactive content to obtain the communication evaluation score of each user; and selecting the user with the highest communication evaluation score among all users as the core user among the users.

[0011] In a second aspect, an embodiment of the present disclosure provides a social media information dissemination evaluation device, which includes: an acquisition unit, configured to acquire users related to current social media information and the relationships between each user; an influence calculation unit, configured to calculate the influence score of each user based on each user and the relationships between each user; an interaction unit, configured to acquire the interactive content of each user related to the current social media information based on the relationship between each user; a quality calculation unit, configured to calculate the quality score of the interactive content of each user based on the interactive content of each user; and an evaluation unit, configured to calculate the dissemination evaluation score of the current social media information based on the influence score of each user and the quality score of the interactive content of each user.

[0012] In some embodiments, the above-mentioned users include: interactive users related to current social media information and associated users related to interactive users; the relationship between the above-mentioned users includes: the relationship between interactive users and current social media information, and the relationship between interactive users and associated users; the above-mentioned influence calculation unit includes: a network building module, which is configured to establish a user relationship network based on interactive users, associated users, the relationship between interactive users and current social media information, and the relationship between interactive users and associated users; a determination module, which is configured to determine the influence score of each user in the user relationship network based on the user relationship network.

[0013] In some embodiments, the interactive content includes: interactive information and recognition; the quality calculation unit includes: an output module configured to input the interactive information of each user into a trained sentiment analysis model to obtain a probability value of the interactive information of each user output by the sentiment analysis model as positive content; an input module configured to input the probability value of the positive content and the recognition into a quality score formula to calculate the quality score of the interactive content of each user; the quality score formula: S X,A =Min(1.0,P X,A -μ+L X,A / ε), where S X,A is the quality score of the interactive content of user X; L X,A is the user X’s recognition of the current social media information A advertisement, L X,A =0 means not approved, L X,A =1 means approval; P X,A ∈[0,1.0] represents the probability value that the interactive information of user X is positive content; μ and ε are specified coefficients.

[0014] In some embodiments, the above-mentioned interactive content includes: recognition; the above-mentioned quality calculation unit is further configured to, for each user, in response to determining that the user has recognition, determine the quality score of the user's interactive content as: 0.5-μ+1 / ε, where μ and ε are specified coefficients.

[0015] In some embodiments, the above-mentioned evaluation unit includes: a multiplication module, configured to multiply the influence score of each user by the quality score of the corresponding interactive content to obtain the dissemination evaluation score of each user; an addition module, configured to add the dissemination evaluation scores of all users to obtain the dissemination evaluation score of the current social media information.

[0016] In some embodiments, the above-mentioned device also includes: a determination unit configured to determine the core users among the users based on the influence scores of the respective users and the quality scores of the interactive contents of the respective users.

[0017] In some embodiments, the above-mentioned determination unit includes: an obtaining module, which is configured to multiply the influence score of each user by the quality score of the corresponding interactive content to obtain the dissemination evaluation score of each user; a ranking module, which is configured to select the user with the highest dissemination evaluation score among all users as the core user among the users.

[0018] In a third aspect, an embodiment of the present disclosure provides an electronic device, comprising: one or more processors; a storage device on which one or more programs are stored; when the one or more programs are executed by one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0019] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any implementation manner in the first aspect.

[0020] The social media information dissemination evaluation method and device provided by the embodiment of the present disclosure first obtains users related to the current social media information and the relationships between each user; then, based on each user and the relationship between each user, calculates the influence score of each user; then, based on the relationship between each user, obtains the interactive content of each user related to the current social media information; then, based on the interactive content of each user, calculates the quality score of the interactive content of each user; finally, based on the influence score of each user and the quality score of the interactive content of each user, calculates the dissemination evaluation score of the current social media information. Thus, by comprehensively considering the information dissemination power and dissemination quality for the dissemination of social media information, it is possible to avoid invalid interactions of fake accounts and inactive accounts, more realistically evaluate the dissemination power of social media advertising, and effectively measure the dissemination effect of social media information. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Other features, objects and advantages of the present disclosure will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0022] Figure 1 is an exemplary system architecture diagram in which an embodiment of the present disclosure may be applied;

[0023] Figure 2 is a flow chart of an embodiment of a method for evaluating the spread of social media information according to the present disclosure;

[0024] Figure 3 is an exemplary schematic diagram of various users and the relationships between various users according to the present disclosure;

[0025] Figure 4 is a flow chart of a method for calculating the influence score of each user according to the present disclosure;

[0026] Figure 5 is a flow chart of a method for calculating the quality score of interactive content of each user according to the present disclosure;

[0027] Figure 6 is a flow chart of a method for calculating a dissemination evaluation score of current social media information according to the present disclosure;

[0028] Figure 7 is a structural schematic diagram of an embodiment of a social media information dissemination evaluation device according to the present disclosure;

[0029] Figure 8 It is a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure. DETAILED DESCRIPTION

[0030] The present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0031] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0032] Figure 1 An exemplary system architecture 100 is shown to which the social media information diffusion evaluation method of the present disclosure can be applied.

[0033] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, and may generally include wireless communication links and the like.

[0034] The terminal devices 101, 102, 103 interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications, such as instant messaging tools, email clients, etc., may be installed on the terminal devices 101, 102, 103.

[0035] The terminal devices 101, 102, 103 can be hardware or software; when the terminal devices 101, 102, 103 are hardware, they can be user devices with communication and control functions, and the above user settings can communicate with the server 105. When the terminal devices 101, 102, 103 are software, they can be installed in the above user devices; the terminal devices 101, 102, 103 can be implemented as multiple software or software modules (for example, software or software modules used to provide distributed services), or they can be implemented as a single software or software module. No specific limitation is made here.

[0036] The server 105 may be a server that provides various services, such as a social media information analysis server that provides support for the social media information analysis system on the terminal devices 101, 102, and 103. The social media information analysis server may analyze and process the relevant information of each user in the network, and feed back the processing results (such as the dissemination evaluation score of the social media information) to the terminal device.

[0037] It should be noted that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, software or software modules used to provide distributed services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0038] It should be noted that the social media information dissemination evaluation method provided in the embodiments of the present disclosure is generally executed by the server 105 .

[0039] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0040] like Figure 2 , shows a process 200 of an embodiment of a social media information dissemination evaluation method according to the present disclosure, the social media information dissemination evaluation method comprising the following steps:

[0041] Step 201: Obtain users related to current social media information and the relationships between users.

[0042] In this embodiment, social media refers to virtual communities and network platforms that people use to create, share, and exchange opinions, views, and experiences. The most significant difference between social media and general social mass media is that social media allows users to enjoy more choices and editing capabilities, and to assemble into a certain reading and listening community. Social media can be presented in a variety of different forms, including text, images, music, and videos. Social media information is information published on the above virtual communities and network platforms, such as advertisements, videos, etc.

[0043] In this embodiment, the user related to the current social media information refers to a valid user who is directly or indirectly related to the current social media information. A valid user refers to a user who has posted at least one content on the social media in a set time period (nearly 3 months) or has at least two fans.

[0044] In this embodiment, the execution subject of the social media information dissemination evaluation method captures valid users directly or indirectly related to each social media information at regular intervals (such as 1 day).

[0045] Furthermore, the execution subject can also use some commonly used crawling tools (such as scrapy, Selenium) to crawl users on social media and the relationships between users. Specifically, when social media is presented in the form of web pages, the crawling tool starts from each user's page, reads its page, obtains the links of their fans or followers and puts them into a queue, continuously obtains links from the queue and repeats the above process until all links have been crawled, thereby obtaining all users related to the current social media information and the relationships between each user.

[0046] Step 202: Calculate the influence score of each user based on each user and the relationship between each user.

[0047] In this embodiment, each user may be a user directly related to the social media information (eg Figure 3 In the example, users 1, 2, and 3 are directly related to the advertisement. These directly related users are also called interactive users. Figure 3 The interactive users are users who have a like relationship D or a comment relationship P with the advertisement; further, each user may also be a user who is further related to the user directly related to the social media information (such as Figure 3 In the example, users 4, 5, and 6 are indirectly related to the advertisement. The indirectly related users are also called associated users. Associated users are users directly related to the interactive users. Figure 3 In the example, user 4 has a fan relationship F with user 1, and user 6 has a follow relationship G with user 1.

[0048] In this embodiment, a user relationship network including all users may be established based on the relationships between the users and all users, and the influence score of each user may be determined according to the ranking of each user in the user relationship network.

[0049] Step 203: based on the relationship between the users, the interactive content of the users related to the current social media information is obtained.

[0050] In this embodiment, the user's feedback on social media information can be of various types, such as the degree of recognition of social media information or interactive information on social media information. The interactive information commenting on social media information can be positive or negative. In this embodiment, the degree of recognition of social media information can be expressed by the user's likes on social media information. For example, if the user likes the social media information, it means that the user approves it; if the user does not like the social media information, it means that the user does not approve it.

[0051] In this embodiment, the interactive information is the information interaction content between each user or multiple users in the process of social media information dissemination. For example, the interactive information includes: comment content and / or forwarded content.

[0052] Specifically, the interactive content includes: interactive information and recognition, or the interactive content includes: recognition, and the recognition includes: expressing recognition and not recognizing.

[0053] Step 204: Calculate the quality score of each user's interactive content based on the interactive content of each user.

[0054] In this embodiment, the interactive content of each user with social media information may be positive, neutral or even negative. For example, "This brand of ice cream is really good" (positive), "How long can ice cream be kept?" (neutral), "Whoever buys this ice cream will regret it" (negative), "There are actually ice cream ads on Weibo now" (partially negative). If there are negative emotions in the comments and forwarded content of the advertisement, then the spread of these contents may harm the brand. Neutral content is relatively good and can help spread moderately. Therefore, it is necessary to analyze all interactive content and evaluate whether each comment / forward has a positive or negative effect.

[0055] In some optional implementations of the present embodiment, the interactive content includes: expressing approval; the above-mentioned calculation of the quality score of each user's interactive content based on the interactive content of each user includes: for each user, in response to determining that the user expresses approval, determining that the quality score of the user's interactive content is: 0.5-μ+1 / ε, wherein μ and ε are specified coefficients, wherein μ and ε can be specified according to empirical values, for example, in one example, ε=5, μ=0.4.

[0056] In this optional implementation, when there is only user approval in the interactive content but no interactive information, the quality score of the interactive content is a fixed value, which ensures the reliability of the subsequent calculation of the communication evaluation score.

[0057] Step 205, based on the influence score of each user and the quality score of the interactive content of each user, calculate the dissemination evaluation score of the current social media information.

[0058] In this embodiment, the influence scores of each user and the quality scores of each user's interactive content calculated in advance can be combined to obtain the score contributed by each user to the current social media information (set as A). The scores of all users who have interacted with A are added together to obtain the total score of the current social media information, which is the dissemination evaluation score of the current social media information.

[0059] In some optional implementations of this embodiment, after step 205, the method further includes: determining core users among the users based on the influence scores of the respective users and the quality scores of the interactive contents of the respective users.

[0060] In this embodiment, core users refer to people who have more and more accurate social media information, are accepted or trusted by the relevant group, and have a greater influence on the purchasing behavior of the group. Based on the influence score of each user and the quality score of the interactive content, the determined core users can be core users with positive evaluation capabilities or core users with negative evaluation capabilities.

[0061] When the core users with positive evaluation ability are determined, a basis can be provided for the dissemination and marketing of social media information. Furthermore, the core users among the determined users can be one core user or multiple core users.

[0062] The method for determining core users provided by this optional implementation determines the core users among users based on the influence score of each user and the quality score of the interactive content of each user after calculating the dissemination evaluation score of the current social media information, thereby providing a basis for determining the core users in the social network and ensuring the reliability when executing marketing strategies.

[0063] Furthermore, when a core user is needed, in some optional implementations of this embodiment, the core user among the users is determined based on the influence score of each user and the quality score of the interactive content of each user, including:

[0064] The influence score of each user is multiplied by the quality score of the corresponding interactive content to obtain the communication evaluation score of each user; the user with the highest communication evaluation score among all users is regarded as the core user among the users.

[0065] The method for determining core users among users provided in this optional embodiment can obtain users with the highest communication evaluation scores among all users, and can effectively determine users who are the center of influence of social media information.

[0066] Optionally, when multiple core users are needed, the core users among the users are determined based on the influence scores of each user and the quality scores of the interactive contents of each user, and the method also includes: multiplying the influence scores of each user by the quality scores of the corresponding interactive contents to obtain the communication evaluation scores of each user; taking the users whose communication evaluation scores are ranked from high to a set position at the bottom as the core users among the users; wherein the set position can be set according to network requirements, for example, the set position is set to 5.

[0067] The social media information dissemination evaluation method provided by the embodiment of the present disclosure first obtains users related to the current social media information and the relationships between each user; then, based on each user and the relationship between each user, calculates the influence score of each user; then, based on the relationship between each user, obtains the interactive content of each user related to the current social media information; then, based on the interactive content of each user, calculates the quality score of the interactive content of each user; finally, based on the influence score of each user and the quality score of the interactive content of each user, calculates the dissemination evaluation score of the current social media information. Thus, by comprehensively considering the information dissemination power and dissemination quality for the dissemination of social media information, it is possible to avoid invalid interactions of fake accounts and inactive accounts, more realistically evaluate the dissemination power of social media advertising, and effectively measure the dissemination effect of social media information.

[0068] In this embodiment, each user includes users directly related to social media information, also called interactive users, and may also include users related to users directly related to social media information, also called associated users.

[0069] In some optional implementations of this embodiment, each user includes: an interactive user related to the current social media information and an associated user related to the interactive user; the relationship between each user includes: the relationship between the interactive user and the current social media information, and the relationship between the interactive user and the associated user. Figure 4 , showing a process 400 of a method for calculating the influence score of each user of the present disclosure, the method for calculating the influence score of each user includes:

[0070] Step 401 , establishing a user relationship network based on interactive users, associated users, the relationship between interactive users and current social media information, and the relationship between interactive users and associated users.

[0071] In this optional implementation, the execution subject captures all interactive users who like, comment, and forward the advertisement, and captures their fans (set as F) and the objects they follow (set as G) until there are no new users to capture. The order of capturing users is as follows: Figure 3 As shown. All captured users and the relationship between their “fans” and “followers” ​​are saved in the database. This process avoids capturing all users of social media (hundreds of millions), and only captures users directly or indirectly related to the advertisement (generally millions to tens of millions, or even smaller).

[0072] The saved valid users and the relationships between valid users are used to build a user relationship network. Specifically, if user A is a fan of user B, a directed edge is connected from A to B. In this way, a user relationship network is built.

[0073] Step 402: Based on the user relationship network, determine the influence score of each user in the user relationship network.

[0074] In this optional implementation, for the constructed user relationship network, the influence score of each user can be determined by calculating the PageRank value (hereinafter referred to as PR value) of each user. The PR value is a component of the search ranking algorithm, with levels ranging from 1 to 10, 10 being the full score. The higher the PR value, the more important the position of the web page in the search ranking. The PR value algorithm is a classic algorithm for web page ranking and will not be described in detail here. In this embodiment, the calculated PR value of each user is saved as the user's communication influence.

[0075] It can be seen from the PR value that having many fans who have no influence and are not very active (and may be "zombie fans") does not increase the KOL's communication influence; on the contrary, high-quality fans, even if there are only a few of them, can make the content published by the KOL go viral. Therefore, KOL users with strong communication capabilities do not necessarily have the most fans, but they must be the center of the social network. Therefore, in this embodiment, the PR value is used as a measurement indicator of user influence.

[0076] Optionally, the influence score of each user can be obtained by modeling, etc. For example, by using deep learning technology, based on the constructed user relationship network of users, the score of each user on the user relationship network output by the trained influence score model is obtained, and the score is the influence score of the user.

[0077] In this optional implementation, interactive users related to social media information are first captured, and then associated users related to the interactive users are captured, thereby avoiding the capture of all users related to social media information. On the basis of effectively capturing users, the pressure on the execution subject in the data processing process is reduced.

[0078] In this embodiment, the user's feedback on social media information can be of various types, such as recognition of social media information, or interactive information on social media comments. The interactive information on social media comments can be positive or negative. In some optional implementations of this embodiment, the interactive content includes: interactive information and recognition;

[0079] See also Figure 5 , shows a process 500 of a method for calculating the quality score of each user's interactive content, the method for calculating the quality score of each user's interactive content includes:

[0080] Step 501: Input the interaction information of each user into the trained sentiment analysis model to obtain the probability value of the interaction information of each user output by the sentiment analysis model being positive content.

[0081] In this optional implementation, the BERT (Bidirectional Encoder Representation from Transformers) model can be used as a sentiment analysis model, which is one of the most advanced natural language processing models. The publicly available Chinese sentiment analysis dataset is used as a transfer learning training sample (such as Chinese Weibo sentiment analysis). Specifically, the training process of the sentiment analysis model includes the following steps:

[0082] 1) From the transfer learning training samples, sample a batch of texts and their corresponding sentiment polarities (positive is 1, negative is 0).

[0083] 2) Encode each text and input it into the BERT model.

[0084] 3) The BERT model outputs the category of a text (set as S) (for example, it can be a 768-dimensional Embedding vector), and uses the category as input to a classifier (which can be implemented using a fully connected layer with a sigmoid activation function). At the same time, the sentiment annotation y (equal to 1 or 0) corresponding to the sampled text S in step 1) is used as the annotation data corresponding to the text to train the classifier.

[0085] Steps 1)-3) are repeated until all samples are trained for K rounds (for example, K is 100), and a trained BERT model is obtained (set as M). Thereafter, the trained BERT model is used to predict the sentiment polarity of each user's interactive content.

[0086] Specifically, taking advertisement A as an example, the process of sentiment analysis model prediction is described in detail as follows:

[0087] For a certain advertisement A, all forwarded and commented content (hereinafter referred to as interactive content) and corresponding users are captured;

[0088] Aggregate multiple interactive contents of each user together (for example, multiple comments P of user 3 in the figure are combined into one comment separated by commas); extract the interactive contents of each user one by one, and assume that the currently extracted user is X; input the interactive content of user X to model M, and the binary classifier of model M will output the probability that the interactive content of user X is positive content, P X,A , where P X,A ∈[0,1.0].

[0089] Step 502: Input the probability value and recognition degree of the positive content into the quality score formula to calculate the quality score of each user's interactive content.

[0090] In this optional implementation, the quality score formula is: S X,A =Min(1.0,P X,A -μ+L X,A / ε), where S X,A is the quality score of the interactive content of user X; L X,A is the user X’s recognition of the current social media information A advertisement, L X,A =0 means not approved, L X,A =1 means approval; P X,A ∈[0,1.0] represents the probability value that the interactive information of user X is positive content; μ and ε are specified coefficients.

[0091] In this embodiment, the calculated S X,A The result is saved in the database as the sentiment score of user X to advertisement A, that is, the quality score of user X to advertisement A.

[0092] In the mass fraction formula, P X,A -μ represents the positive or negative sentiment of the user interaction content, while L X,A / ε is used to measure the impact of user approval (such as likes).

[0093] If P X,A >0.5, then user X is more likely to interact with ad A, otherwise it is negative. X,A Content that is closer to 0.5 (for example, greater than 0.4 and less than 0.5) is generally more neutral (not obviously negative), and its dissemination is still valuable, so a relaxation operation can be performed to make μ equal to 0.4 instead of 0.5.

[0094] If (P X,A -μ)<0, but L X,A =1, that is, although the user's interactive content is negative, he liked it, which is a contradiction. There are two possibilities. One is that the sentiment binary classification of the BERT model is inaccurate in this case, and the user's interactive content is at least neutral; the other is that the user did post negative content, but mistakenly liked it. The spread of written content is far greater than the user's like behavior. Therefore, the ε variable is used in the formula to weaken the impact of likes (in this example, ε = 5, that is, likes increase the sentiment score by 0.2).

[0095] From another perspective, the time and energy spent by users in writing content (even if it is just a few simple words) is far greater than the act of liking. Therefore, behaviors such as forwarding and commenting can better reflect the user's emotional attitude. Generally speaking, ε should be greater than 2 to be more reasonable.

[0096] This embodiment can identify fake users and low-activity users on social media, and reasonably evaluate the marketing effects on them, avoiding paying excessive advertising fees for low-value and worthless users.

[0097] The method for calculating the quality score of interactive content provided by this optional implementation, for interactive content including interactive information and recognition, inputs the interactive information of each user into a sentiment analysis model to obtain a probability value that the interactive information is positive content; inputs the probability value of the positive content and the recognition into a quality score formula to obtain the quality score of the interactive content. Therefore, the method for calculating the quality score of interactive content, that is, calculating the probability value of the positivity of the interactive information, ensures the reliability of the quality score calculation of the interactive content by adding the recognition into the calculation process.

[0098] In this embodiment, to evaluate the dissemination effect of social media information, it is necessary to comprehensively consider the user's influence score and the quality score of the user's interactive content.

[0099] See also Figure 6 , showing a process 600 of a method for calculating a dissemination evaluation score of current social media information of the present disclosure, the method for calculating a dissemination evaluation score of current social media information includes:

[0100] Step 601, multiplying the influence score of each user by the quality score of the corresponding interactive content to obtain the communication evaluation score of each user.

[0101] In this optional implementation, taking the current advertisement A as an example, all users who like, comment, and forward it are captured, and the user set is set as U A . A For each user u in the table, generate an influence score table including the influence scores of each user and a sentiment score table including the quality scores of each user, and take out the influence score PR of user u u , and the quality score S of user u for ad A u,A .

[0102] Calculate the propagation evaluation score R of user u u,A , communication evaluation score = influence score * quality score, that is, R u,A =PR u *S u,A .

[0103] Step 602: Add the dissemination evaluation scores of all users to obtain the dissemination evaluation score of the current social media information.

[0104] In this optional implementation, the communication effects of all users who interact with advertisement A are added together to obtain its total communication effect score, which is the communication evaluation score R of advertisement A. A ,Right now

[0105] Through the above steps, we can get the communication evaluation score R of the current social media advertisement A. A , which is the marketing effectiveness score of the advertisement on social media.

[0106] The present invention also integrates the communication quality score to identify advertisements with negative user sentiment. u,A =PR u *S u,A If users are more disgusted with the advertisement, then the advertising influence (PR u ) is larger, the lower the marketing effectiveness score is.

[0107] This optional implementation method provides a method for calculating the dissemination evaluation score of the current social media information. The dissemination evaluation score of each user is first calculated, and then the dissemination evaluation scores of all users are added together to obtain the dissemination evaluation score of the current social media information. This ensures the effectiveness of calculating the dissemination evaluation score of the current social media information based on the comprehensive effects of the dissemination evaluation scores of all users.

[0108] Further references Figure 7 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a social media information dissemination evaluation device. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0109] like Figure 7As shown, an embodiment of the present disclosure provides a social media information dissemination evaluation device 700, and the device 700 includes: an acquisition unit 701, an influence calculation unit 702, an interaction unit 703, a quality calculation unit 704 and an evaluation unit 705. Among them, the acquisition unit 701 can be configured to acquire users related to the current social media information and the relationship between each user. The influence calculation unit 702 can be configured to calculate the influence score of each user based on each user and the relationship between each user. The interaction unit 703 can be configured to acquire the interactive content of each user related to the current social media information based on the relationship between each user. The quality calculation unit 704 can be configured to calculate the quality score of the interactive content of each user based on the interactive content of each user. The evaluation unit 705 can be configured to calculate the dissemination evaluation score of the current social media information based on the influence score of each user and the quality score of the interactive content of each user.

[0110] In this embodiment, the specific processing of the acquisition unit 701, the influence calculation unit 702, the interaction unit 703, the quality calculation unit 704 and the evaluation unit 705 and the technical effects thereof can be referred to in detail. Figure 2 Corresponding to step 201, step 202, step 203, step 204, and step 205 in the embodiment.

[0111] In some embodiments, the above-mentioned users include: interactive users related to the current social media information and associated users related to the interactive users; the relationship between the above-mentioned users includes: the relationship between the interactive users and the current social media information, and the relationship between the interactive users and the associated users; the above-mentioned influence calculation unit 702 includes: a network building module (not shown in the figure) and a determination module (not shown in the figure). Among them, the network building module can be configured to establish a user relationship network based on the interactive users, the associated users, the relationship between the interactive users and the current social media information, and the relationship between the interactive users and the associated users. The determination module can be configured to determine the influence score of each user in the user relationship network based on the user relationship network.

[0112] In some embodiments, the above-mentioned interactive content includes: interactive information and recognition; the above-mentioned quality calculation unit 704 includes: an output module (not shown in the figure) and an input module (not shown in the figure). The output module can be configured to input the interactive information of each user into a trained sentiment analysis model to obtain the probability value of each user's interactive information output by the sentiment analysis model as positive content. The input module can be configured to input the probability value of positive content and recognition into the quality score formula to calculate the quality score of each user's interactive content; the quality score formula is: S X,A=Min(1.0,P X,A -μ+L X,A / ε), where S X,A is the quality score of the interactive content of user X; L X,A is the user X’s recognition of the current social media information A advertisement, L X,A =0 means not approved, L X,A =1 means approval; P X,A ∈[0,1.0] represents the probability value that the interactive information of user X is positive content; μ and ε are specified coefficients.

[0113] In some embodiments, the above-mentioned interactive content includes: recognition; the above-mentioned quality calculation unit 704 is further configured to, for each user, in response to determining that the user has recognition, determine the quality score of the user's interactive content as: 0.5-μ+1 / ε, where μ and ε are specified coefficients.

[0114] In some embodiments, the evaluation unit 705 includes: a multiplication module (not shown in the figure) and an addition module (not shown in the figure). The multiplication module can be configured to multiply the influence score of each user by the quality score of the corresponding interactive content to obtain the communication evaluation score of each user. The addition module can be configured to add the communication evaluation scores of all users to obtain the communication evaluation score of the current social media information.

[0115] In some embodiments, the apparatus 700 further includes: a determination unit (not shown in the figure). The determination unit may be configured to determine core users among the users based on the influence scores of the users and the quality scores of the interactive contents of the users.

[0116] In some embodiments, the determination unit includes: an acquisition module (not shown in the figure) and a ranking module (not shown in the figure). The acquisition module can be configured to multiply the influence score of each user by the quality score of the corresponding interactive content to obtain the communication evaluation score of each user. The ranking module can be configured to select the user with the highest communication evaluation score among all users as the core user among the users.

[0117] The social media information dissemination evaluation device provided by the embodiment of the present disclosure first obtains the users related to the current social media information and the relationship between each user; then, the influence calculation unit 702 calculates the influence score of each user based on the relationship between each user; then, the interaction unit 703 obtains the interactive content related to the current social media information of each user based on the relationship between each user; then, the quality calculation unit 704 calculates the quality score of the interactive content of each user based on the interactive content of each user; finally, the evaluation unit 705 calculates the dissemination evaluation score of the current social media information based on the influence score of each user and the quality score of the interactive content of each user. Therefore, the information dissemination power and dissemination quality are comprehensively considered for the dissemination of social media information, which can avoid the invalid interaction of false accounts and inactive accounts, and more realistically evaluate the dissemination power of social media advertising, thereby effectively measuring the dissemination effect of social media information.

[0118] Reference below Figure 8 , which shows a structural schematic diagram of an electronic device 800 suitable for implementing an embodiment of the present disclosure.

[0119] like Figure 8 As shown, the electronic device 800 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage device 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the electronic device 800 are also stored. The processing device 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0120] Typically, the following devices may be connected to the I / O interface 805: input devices 806 including, for example, a touch screen, a touch pad, a keyboard, a mouse, etc.; output devices 807 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 808 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 809. The communication device 809 may allow the electronic device 800 to communicate with other devices wirelessly or by wire to exchange data. Figure 8 The electronic device 800 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 8 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0121] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 809, or installed from a storage device 808, or installed from a ROM 802. When the computer program is executed by the processing device 801, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0122] It should be noted that the computer-readable medium of the embodiment of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the embodiment of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the embodiment of the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0123] The computer-readable medium may be included in the server; or it may exist independently without being installed in the server. The computer-readable medium carries one or more programs. When the one or more programs are executed by the server, the server: obtains users related to the current social media information and the relationships between each user; calculates the influence score of each user based on the users and the relationships between each user; obtains the interactive content related to the current social media information of each user based on the relationships between each user; calculates the quality score of the interactive content of each user based on the interactive content of each user; calculates the dissemination evaluation score of the current social media information based on the influence score of each user and the quality score of the interactive content of each user.

[0124] Computer program code for performing the operations of embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on a user's computer, partially on a user's computer, as a separate software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0125] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0126] The units involved in the embodiments described in the present disclosure may be implemented by software or by hardware. The described units may also be provided in a processor, for example, may be described as: a processor comprising an acquisition unit, an influence calculation unit, an interaction unit, a quality calculation unit, and an evaluation unit. The names of these units do not, in some cases, constitute limitations on the units themselves, for example, the acquisition unit may also be described as a unit "configured to obtain users related to the current social media information and the relationships between the users".

[0127] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) to form a technical solution.

Claims

1. A method for evaluating the dissemination of information on social media, the method comprising: Acquire users related to the current social media information and the relationships between the users, wherein the users include interactive users and associated users of the current social media information, and the associated users are users that can be traced back to the interactive users based on fan relationships and / or follow relationships; Establishing a user relationship network based on the interactive user, the associated user, the relationship between the interactive user and the current social media information, the relationship between the interactive user and the associated user, and the relationship between the associated users; Based on the user relationship network, calculate the PageRank value of each user; Determine the influence score of each user based on their PageRank value; Based on the relationship between each user, obtain the interactive content of each user related to the current social media information; Based on the interactive content of each user, the quality score of the interactive content of each user is calculated; Based on the influence score of each user and the quality score of each user's interactive content, the dissemination evaluation score of the current social media information is calculated.

2. The method according to claim 1, wherein: The interactive content includes: interactive information and recognition; the quality score of each user's interactive content is calculated based on the interactive content of each user, including: Inputting the interaction information of each user into the trained sentiment analysis model to obtain a probability value of the interaction information of each user output by the sentiment analysis model being positive content; Input the probability value and recognition degree of the positive content into the quality score formula to calculate the quality score of each user's interactive content; The mass fraction formula: S X,A =Min(1.0,P X,A -μ+L X,A / ε), where S X,A is the quality score of the interactive content of user X; L X,A is the user X’s recognition of the current social media information A advertisement, L X,A =0 means not approved, L X,A =1 means approval; P X,A ∈[0,1.0] represents the probability value that the interactive information of user X is positive content; μ and ε are specified coefficients.

3. The method according to claim 1, wherein: The interactive content includes: expressing recognition; the quality score of each user's interactive content is calculated based on the interactive content of each user, including: For each user, in response to determining that the user has expressed approval, the quality score of the interactive content of the user is determined to be: 0.5-μ+1 / ε, where μ and ε are specified coefficients.

4. The method according to claim 1, wherein: The calculation of the current social media information dissemination evaluation score based on the influence score of each user and the quality score of the interactive content of each user includes: Multiply the influence score of each user by the quality score of the corresponding interactive content to obtain the communication evaluation score of each user; Add up the dissemination evaluation scores of all users to obtain the dissemination evaluation score of the current social media information.

5. The method according to any one of claims 1 to 4, further comprising: Identify core users based on the influence score of each user and the quality score of each user's interactive content.

6. The method according to claim 5, wherein: Determining the core users among the users based on the influence scores of the users and the quality scores of the interactive contents of the users includes: Multiply the influence score of each user by the quality score of the corresponding interactive content to obtain the communication evaluation score of each user; The user with the highest propagation evaluation score among all users is regarded as the core user among the users.

7. A social media information dissemination evaluation device, the device comprising: an acquisition unit, configured to acquire users related to the current social media information and the relationships between the users, wherein the users include interactive users and associated users of the current social media information, and the associated users are users that can be traced back to the interactive users based on fan relationships and / or follow relationships; The influence calculation unit is configured to establish a user relationship network based on the interactive user, the associated user, the relationship between the interactive user and the current social media information, the relationship between the interactive user and the associated user, and the relationship between the associated users; calculate the PageRank value of each user based on the user relationship network; determine the influence score of each user based on the PageRank value of each user; An interaction unit, configured to obtain interaction content related to current social media information of each user based on the relationship between each user; A quality calculation unit, configured to calculate a quality score of each user's interactive content based on the interactive content of each user; The evaluation unit is configured to calculate the dissemination evaluation score of the current social media information based on the influence score of each user and the quality score of the interactive content of each user.

8. The device according to claim 7, wherein: The interactive content includes: interactive information and recognition; the quality calculation unit includes: An output module is configured to input the interaction information of each user into a trained sentiment analysis model to obtain a probability value that the interaction information of each user output by the sentiment analysis model is positive content; An input module is configured to input the probability value and recognition degree of the positive content into a first quality score formula to calculate the quality score of the interactive content of each user; The mass fraction formula: S X,A =Min(1.0,P X,A -μ+L X,A / ε), where S X,A is the quality score of the interactive content of user X; L X,A is the user X’s recognition of the current social media information A advertisement, L X,A =0 means not approved, L X,A =1 means approval; P X,A ∈[0,1.0] represents the probability value that the interactive information of user X is positive content; μ and ε are specified coefficients.

9. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

10. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images