Construction Of Wechat Information Quality Evaluation Index System

WeChat has become the main representative of mobile social media,which nowadays occupies an important part of people's life.WeChat mainly builds a circle of friends through acquaintances.The content of WeChat information quality evaluation is not the interactive information between friends,but the various marketing and promotional information generated by WeChat platform.Since WeChat Accounts launched in 2012,it has quickly become an important commercial promotion channel.So how to help users to filter out valuable information has become an urgent problem to be solved.In addition,a large number of WeChat accounts output a large number of articles every day,making it more difficult for users to obtain their required information.The main research content of this article is: analyze the theory and methods of social media and WeChat information quality evaluation,then extract the information quality indicators applicable to the evaluation of WeChat Accounts articles based on the current evaluation index of information quality and the information features of WeChat.After that,analyze the availability and feasibility of indicators by literature research and empirical research to ensure the operability and applicability of the indicators and further verify the scientificity and rationality of the index through machine learning.Based on the above research,the WeChat information quality evaluation index system constructed in this paper includes three target level indicators(public platform,article content and user perspective),8 criteria indicators and 13 measurement indicators.The criteria level indicators of the public platform are: authority,influence,and originality;criterion level indicators of the content of article are: relevance,reliability,originality,and attractiveness;criterion level indicators of the user's perspective are: perceived usefulness.The 13 measures are as follows: The authoritativeness of the public platform includes certification;the influence includes the WCI index,the number of active fans,the number of releases,and the number of published articles;the originality is represented by whether the public platform is an original platform.The reliability of the content of the article includes whether there is a reference source of information,whether it contains advertisements;originality includes whether the article is original;relevance measures whether the article subject is related to the title;attraction is whether the article contains pictures or videos.Perceived usefulness includes readings and likes.This article takes the corporate public WeChat in the health field as an example,learns and trains the samples through the Decision Tree C5.0.The accuracy rate of the verification result obtained is 83.97%,which shows that the evaluation index system proposed in this paper has good applicability and accuracy.