Research On Evaluation Of Answer Quality Of Online Q&A Site

The intensification of modern people’s knowledge anxiety symptoms and the rapid spread of the Internet have provided broad development prospects for online question and answer platforms.The mode of multi-person collaborative editing of the same question by gathering group wisdom has been widely welcomed by contemporary young people.However,not all online Q&A platforms can survive the fierce market competition.If users’ Q&A needs are not well met,it will be a matter of time before the loss of users and the decline of the platform.Answers are an important asset of online question and answer platforms and the key to attracting users to the platform.Answer quality management is the focus of the daily work of online question and answer platform managers.But at the same time,answering is a double-edged sword.High-quality answers can increase the user stickiness of the platform.Low-quality answers will cause the loss of platform users.A question-and-answer platform will naturally lose a large number of users and cannot survive.At present,the answer quality problems existing in online Q&A platforms mainly include false answers,redundant answers,outdated answers,and false answers.Although the platform has formulated a series of answer quality review plans to purify the platform’s question and answer environment,there is no effective answer quality evaluation mechanism,and everything is a temporary solution instead of a root cause.Through literature research,it is found that there are relatively few researches on answer quality evaluation of online question and answer platforms at home and abroad,which are mainly divided into three aspects: research on answer quality influencing factors,research on manual evaluation of answer quality,and research on automated evaluation of answer quality.Although these studies have different methods and names when extracting evaluation indicators,there are common points in different indicator systems,and these indicators can be summarized and summarized.Answer quality is a complex evaluation object.Because it is greatly affected by human subjective factors,it is very difficult to evaluate it,and many factors are difficult to quantify.This article uses the fuzzy comprehensive evaluation method,uses fuzzy mathematics to quantify the various factors that affect the quality of the answer,and establishes a fuzzy set.The specific method is to summarize the evaluation indicators used by predecessors in the literature research,and according to the communication system model,these indicators are based on the source of the answer and the content of the answer.,External factors and response utility are divided into four categories.The source of the answers includes the professionalism,originality and attitude of the answers;the content of the answers includes the completeness,accuracy,relevance,legibility,objectivity and vividness of the answers,and the external factors include Answers can be verified,external links,and answer efficiency.Answer utility includes three factors: response timeliness,feasibility and appropriateness.Use these indicators as a set of factors,take good,average,and poor as a set of comments,use the analytic hierarchy process to determine the relative weight of each factor in the set of factors,combine the results of the evaluation table to construct a fuzzy evaluation matrix for fuzzy synthesis,and normalize the results After processing,you can get a rating of the quality of the answer,whether the quality of the answer is good,fair or bad.Finally,it is an empirical analysis.The three answers ranked first,middle,and second under the topic of "new crown vaccination" on the Zhihu Q&A platform are selected as sample answers,and the quality evaluation is made using the constructed fuzzy comprehensive evaluation model of answer quality.The evaluation results are in line with reality.The situation shows that the comprehensive evaluation model of answer quality of the online question and answer platform constructed in this article can accurately judge the quality of answers and help the platform to screen answers in advance.