CN-121616282-B - Affinity evaluation method and device in social network and electronic equipment

CN121616282BCN 121616282 BCN121616282 BCN 121616282BCN-121616282-B

Abstract

The application provides a method, a device and electronic equipment for evaluating intimacy in a social network, and relates to the technical field of Internet. The method comprises the steps of evaluating first interaction behaviors from three dimensions of comment continuity, comment timeliness and comment influence, determining a first interaction score representing interaction liveness between a target user and a target creator, evaluating second interaction behaviors from two dimensions of forwarding timeliness and forwarding frequency, determining a second interaction score representing propagation influence of the target user on user generated content issued by the target creator, and determining the intimacy between the target user and the target creator based on the first interaction score and the second interaction score. By adopting the method and the device for evaluating the intimacy in the social network and the electronic equipment, accuracy and reliability of intimacy calculation results are improved.

Inventors

ZHANG WEISHI
QI CHAOYANG
MENG XINGYU
WANG CHUNXIA
LI WEIHAO
LI XIAOYE
SONG ZHEMING
LIU TONGJU
YAN SIRUI
SU ZHENYU
GAO YU
CHEN BO

Assignees

中国电子信息产业集团有限公司第六研究所

Dates

Publication Date: 20260508
Application Date: 20251204

Claims (8)

1. A method for affinity evaluation in a social network, comprising: Determining a target user generating interaction behaviors with a target creator based on social network data, wherein the interaction behaviors comprise a first interaction behavior for commenting user generated content and a second interaction behavior for forwarding user generated content; Evaluating the first interaction behavior from three dimensions of comment continuity, comment timeliness and comment influence, and determining a first interaction score representing interaction liveness between the target user and the target creator; Evaluating the second interaction behavior from two dimensions of forwarding timeliness and forwarding frequency, and determining a second interaction score representing the propagation influence of the target user on the user generated content issued by the target creator; Determining an affinity between the target user and the target creator based on the first interaction score and the second interaction score; the first interaction behavior comprises at least one comment of the user-generated content, the first interaction score comprises a comment score corresponding to the comment, and the first interaction behavior is evaluated by the following method: For each comment, evaluating the comment continuity of the first interaction behavior under the comment based on the number of continuous comments when the comment is stopped, so as to determine a continuity score; Based on the comment time interval between the comment and the previous comment and the distance between the comment and the current moment, evaluating the comment timeliness of the first interactive behavior under the comment so as to determine a comment timeliness score; Based on the praise number and the return number of the comment, evaluating the comment influence of the first interaction behavior under the comment to determine an influence score; Determining a comment score of the first interaction behavior under the comment based on the continuity score, the comment timeliness score and the influence score; The second interaction behavior comprises at least one forwarding of generated content for different users, the second interaction behavior being evaluated by: based on the time interval of each forwarding distance from the current moment, evaluating the forwarding timeliness of the second interaction behavior to determine a forwarding timeliness score; based on the forwarding times and the forwarding times of all target users, evaluating the forwarding frequency of the second interaction behavior to determine a forwarding frequency score; And determining a second interaction score based on the forwarding timeliness score and the forwarding frequency score.
2. The method of claim 1, wherein a first interaction score characterizing an activity of interaction between the target user and the target creator is determined by: determining a comment total score based on the sum of comment scores of the first interactive behaviors under each comment; And determining the heat degree of the user generated content based on the total score of the user generated content, and determining a first interaction score based on the heat degree and the total score of the score.
3. The method of claim 1, wherein the comment score for the first interactive behavior under the present comment is determined by: Determining a base comment score based on a total number of comments of the target user for the user-generated content; Comparing the interaction timeliness score corresponding to the continuity score and the comment timeliness score with a set weight threshold to determine interaction weight based on a comparison result; And determining the comment score of the first interaction behavior under the comment based on the basic comment score, the interaction weight and the influence score.
4. The method of claim 1, wherein evaluating the forwarding frequency of the second interaction behavior based on the forwarding times and the forwarding times of all target users to determine a forwarding frequency score comprises: Selecting the forwarding frequency with the largest value from the total forwarding frequency of candidate users as target forwarding frequency, wherein the candidate users are users forwarding user generated content issued by target creators; And determining a forwarding frequency score based on the forwarding times of the target user and the target forwarding times.
5. The method of claim 1, wherein evaluating the forwarding timeliness of the second interaction behavior based on the time interval of each forwarding from the current time to determine a forwarding timeliness score comprises: Determining a single forwarding timeliness score based on the time interval power of the natural constant; and determining the forwarding timeliness score based on the average value of all the single forwarding timeliness scores.
6. The method of claim 1, wherein the target user comprises a plurality, the method further comprising: and sequencing the plurality of target users according to the order of the intimacy from high to low so as to determine the intimacy of different target users to the target creator.
7. An affinity evaluation device in a social network, comprising: The selection module is used for determining a target user generating interaction behaviors with a target creator based on social network data, wherein the interaction behaviors comprise a first interaction behavior for commenting user generated content and a second interaction behavior for forwarding the user generated content; The first evaluation module is used for evaluating the first interaction behavior from three dimensions of comment continuity, comment timeliness and comment influence to determine a first interaction score representing interaction liveness between the target user and the target creator; The second evaluation module is used for evaluating the second interaction behavior from two dimensions of forwarding timeliness and forwarding frequency and determining a second interaction score representing the propagation influence of the target user on the user generated content issued by the target creator; The comprehensive evaluation module is used for determining the intimacy between the target user and the target creator based on the first interaction score and the second interaction score; the first interaction behavior comprises at least one comment aiming at the user generated content, the first interaction score comprises a comment score corresponding to the comment, and the first evaluation module is specifically configured to: For each comment, evaluating the comment continuity of the first interaction behavior under the comment based on the number of continuous comments when the comment is stopped, so as to determine a continuity score; Based on the comment time interval between the comment and the previous comment and the distance between the comment and the current moment, evaluating the comment timeliness of the first interactive behavior under the comment so as to determine a comment timeliness score; Based on the praise number and the return number of the comment, evaluating the comment influence of the first interaction behavior under the comment to determine an influence score; Determining a comment score of the first interaction behavior under the comment based on the continuity score, the comment timeliness score and the influence score; the second interaction behavior comprises at least one forwarding of content generated for different users, and the second evaluation module is specifically configured to: based on the time interval of each forwarding distance from the current moment, evaluating the forwarding timeliness of the second interaction behavior to determine a forwarding timeliness score; based on the forwarding times and the forwarding times of all target users, evaluating the forwarding frequency of the second interaction behavior to determine a forwarding frequency score; And determining a second interaction score based on the forwarding timeliness score and the forwarding frequency score.
8. An electronic device comprising a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor in communication with the memory via the bus when the electronic device is operating, the processor executing the machine-readable instructions to perform the steps of the affinity assessment method in a social network as claimed in any one of claims 1 to 6.

Description

Affinity evaluation method and device in social network and electronic equipment Technical Field The application relates to the technical field of internet, in particular to a method and a device for evaluating intimacy in a social network and electronic equipment. Background With the rapid development of internet technology, the intersection of people's work and life with the internet is becoming more and more compact, and social networks have become an integral part of people's lives. In practical application, users can add other users with interests as friends through the social application platform to release user generated contents such as a tweet, a video and the like, and can also perform interactive operations such as praise, comment and the like on the user generated contents released by the friends, so that communication among users is greatly facilitated. How to effectively expand and maintain the relationship among users and mine out the intimacy among users is a core problem which needs to be solved by each social network platform. However, in the prior art, the intimacy between users is usually calculated based on the reading amount and the praise number of the work published by the social account, which reduces the accuracy and reliability of the intimacy calculation result. Disclosure of Invention Accordingly, an object of the present application is to provide a method and apparatus for evaluating affinity in a social network, and an electronic device, so as to overcome at least one of the above drawbacks. In a first aspect, an embodiment of the present application provides a method for evaluating affinity in a social network, including: determining a target user generating interaction behaviors with a target creator based on social network data, wherein the interaction behaviors comprise a first interaction behavior for commenting user generated content and a second interaction behavior for forwarding user generated content; evaluating the first interaction behavior from three dimensions of comment continuity, comment timeliness and comment influence, and determining a first interaction score representing interaction liveness between a target user and a target creator; evaluating a second interaction behavior from two dimensions of forwarding timeliness and forwarding frequency, and determining a second interaction score representing the propagation influence of a target user on user generated content issued by a target creator; Based on the first interaction score and the second interaction score, an affinity between the target user and the target creator is determined. In an alternative embodiment, the first interaction behavior comprises at least one comment for user generated content, the first interaction score comprises a comment score corresponding to the comment, the first interaction behavior is evaluated by evaluating, for each comment, comment continuity of the first interaction behavior under the comment based on the number of continuous comments when the comment is cut off to the comment, so as to determine a continuity score, evaluating comment timeliness of the first interaction behavior under the comment based on a comment time interval between the comment and the last comment and the distance between the comment and the current moment, so as to determine a comment timeliness score, evaluating comment influence of the first interaction behavior under the comment based on the praise number and the reply number of the comment, so as to determine an influence score, and determining the comment timeliness score of the first interaction behavior under the comment based on the continuity score, the comment timeliness score and the influence score. In an alternative embodiment, the second interaction behavior comprises generating at least one forwarding of the content for different users, the second interaction behavior being evaluated by evaluating a forwarding timeliness of the second interaction behavior based on a time interval of each forwarding from a current time to determine a forwarding timeliness score, evaluating forwarding frequencies of the second interaction behavior based on the forwarding times and the forwarding times of all target users to determine a forwarding frequency score, and determining the second interaction score based on the forwarding timeliness score and the forwarding frequency score. In an alternative embodiment, a first interaction score that characterizes the activity of interactions between the target user and the target creator is determined by determining a total score of comments based on a sum of the comment scores of the first interaction behavior under each comment, determining a popularity of the user-generated content based on the total number of comments of the user-generated content, and determining the first interaction score based on the popularity and the total score of comments. In an alternative embodiment, the comment score of the