Incentivized Comment Detection with Sentiment Analysis on Online Hotel Reviews

With the enormous platforms available in present days, consumers communicate and interconnect online with web users all around the world to share their experiences. Thus, online platform has become a major source of reviews about different entities. People presently travel frequently around the world for different purposes. Seeking good hotels for accommodation is a prime concern. Customer reviews on hotels help future customers to take decisions about their accommodation as well as help hotel owners to rethink about designing customer facilities. However, many online reviews are biased due to different factors. Many hotel owners come up with attractions like referral rewards, coupons, bonus points etc. to the reviewers to motivate them in writing biased reviews. We have worked on US’s 100 hotel and found 952 incentivized reviews out of 19175 reviews, which is 4.96% of total reviews. A categorization on incentivized reviews is performed as well. Furthermore, hotels are distinguished based on real and incentivized reviews found on them. Results are verified using machine learning algorithms. Random Forest, K-Nearest Neighbor and Support Vector Machine are applied as machine learning algorithms to validate the accuracy of our model and their prediction results are compared. Random Forest outperforms with 94.4% prediction accuracy.