The Influence Of Product Perception From Reviews To Consumer Online Behaviors

Online reviews are a big treasure for platforms,manufacture and potential consumers.The number of studies focused on mining online reviews’ value increases sharply recently.Most of them concentrate on reviews separately one by one to find some most helpful reviews for consumers.However,experience of shopping online indicates reading information is a continuous process to update perception of product uncertainty.Thus,our objective is exploring the relations between product uncertainty and consumer behaviors(whether purchase product or vote a view helpful).We improve the Shannon entropy to measure product uncertainly from textual contents(description and reviews).By using real data from Amazon.com,empirical results demonstrate that for search goods rating only connects with negative emotion of reviews textual contents.Reduced uncertainty by a review is closely related to helpful vote no matter all reviews ordered by time sequence or helpful vote.Only the uncertainty of top 20 helpful reviews has effect on product rank for search goods,while for experience goods,the uncertainty of total reviews,top 20 helpful reviews and other conditions show effect on product rank.However,unlike the intuition that consumers would like to vote for reviews which help them decrease product uncertainty,the data analysis show opposite results that they prefer to vote reviews with high varied uncertainty.Besides,we also find that the less product uncertainty measured from reviews,the high sale rank of the product is.Additionally,price affects not only product rank,but also helpful vote.To some extent the effect of product type to helpful votes,is amplified in previous studies without regard to price.Finally,we also discuss the implications for both literature and practice combining with theories of behavioral psychology.