Research On The Influential Factors Of The Usefulness Of Online Reviews That Incorporate Emotional Tendencies

With the rapid development of e-commerce,online shopping has been closely linked with our life.However,the consumers always would be affected by the virtual goods and uncertainties of online goods so that they fail to understand the real information of the goods.It causes a number of unscrupulous merchants create a large number of online reviews on goods through some fake means such as hired individuals or favorable returns.These actions virtually increase the time for consumers to obtain and filter out available information.Therefore,how to really extract the useful information among a large number of reviews has attracted much attention from the majority of scholars.This article utilizes results from relevant research fields and conducts research from the perspective of the usefulness of online reviews.The main tasks of this paper are as follow:(1)A review of the influencing factors over usefulness of online reviews: First,the background,issues,and objectives of the study are introduced;Then,the study of the factors affecting usefulness in the current field is discussed,and the results and existing problems in the comprehensive study are presented,and the content of the study is proposed.(2)Influencing factors extraction and model construction: we extract influencing factors from the perspectives of review features and reviewer characteristics,studying the scores,useful and total number of votes,review length and sentences,the time interval and readability of the review,and reviewer affectiveness influences the usefulness of online reviews.Then,hypotheses are proposed and the hierarchical equation model is constructed.(3)Model validation analysis: On the one hand,we use Tobit regression analysis to test the hypothesis and predict the influence coefficient of the dependent variable;On the other hand,the classification prediction performance of the model was tested by the recall rate,precision rate,and comprehensive index F,and the contribution rate of the PCA was used to rank the importance of the influencing factors.The results of the model designed in this paper can provide new ideas for further research on whether or not the review is useful.It would provide support for consumers to reduce the time cost of browsing reviews and make quick purchase decisions as well.