An Intelligent Approach to Review Filtering and Review Quality Improvement
Iain Lee, Yu Sun, Yuexin Li · 2016
This paper presents a new approach to review filtering in order to bring a more transparent solution to generate more trust between the user base and review-based sites. Instead of removing reviews based on authenticity, users decide on the level of filtering that is provided on the reviews. Each review is given a score and this scoring system is based on three categories: location based check-in (LBS), receipt authenticity, and sentiment analysis on the actual review. Once all three categories are factored, an algorithm will be used to return a score on the review. This score determines the supposed authenticity (confidence that the review is a legitimate review) of the review and this score is what will be used as the filtering mechanism for the user.