Voting for Deceptive Opinion Spam Detection

摘要

Consumers' purchase decisions are increasingly influenced by user-generatedonline reviews. Accordingly, there has been growing concern about the potentialfor posting deceptive opinion spam fictitious reviews that have beendeliberately written to sound authentic, to deceive the readers. Existingapproaches mainly focus on developing automatic supervised learning basedmethods to help users identify deceptive opinion spams. This work, we used the LSI and Sprinkled LSI technique to reduce thedimension for deception detection. We make our contribution to demonstrate whatLSI is capturing in latent semantic space and reveal how deceptive opinions canbe recognized automatically from truthful opinions. Finally, we proposed avoting scheme which integrates different approaches to further improve theclassification performance.