Research Of Online Reviews' Usefulness Based On Solving Negative Word Of Mouth

With the development of user generated content(UGC), online negative word of mouth(ONWM) bring loss to enterprises. How to rapid process massive ONWM become an urgently problem for enterprises and resources recognition is the key of automation solutions. The main purposes of mouth of solving ONWM are relieving complainers' negative emotion and dealing with complain problems. From these two angles, useful online reviews can be the resource of solving ONWM, building the useful model can better identify these resources.Based on emotion and information dimensions for solving ONWM, this paper builds online reviews usefulness recognition model from emotional feature(emotional polarity and emotional intensity), text feature(brand description and product description), form feature(publisher features, pictures, reviews' length). By empirical method, this paper uses mobile phone reviews form BBS to verify the model and calculate coefficient. Results show that(1) Emotional polarity has positive comments on reviews' usefulness, and emotion intensity has negative impact.(2) Different formal features all have positive effect on reviews' usefulness.(3) Various product features have positive influence on reviews' usefulness, but influence of brand description for reviews' usefulness is non-significant. The results meet the emotion and information needed for solving ONWM and can be applied to resources recognition. The existing researches related to online reviews rarely applied to solving ONWM, this paper make up this shortage. This paper use TFIDF to calculate indexes, more accurate than scoring or artificial judging method which applied by most other researches. At the same time, this paper dig the value of online reviews, provide advice and reference for enterprises to identify resources for solving ONWM.