PROJECT TITLE:

Rating Prediction Based On Social Sentiment from Textual Reviews - 2016

ABSTRACT:

In recent years, we tend to have witnessed a flourish of review websites. It presents a great chance to share our viewpoints for various merchandise we tend to purchase. However, we face an info overloading downside. How to mine valuable information from reviews to perceive a user's preferences and create an accurate recommendation is crucial. Traditional recommender systems (RS) take into account some factors, such as user's purchase records, product class, and geographic location. In this work, we tend to propose a sentiment-based mostly rating prediction methodology (RPS) to enhance prediction accuracy in recommender systems. Firstly, we tend to propose a social user sentimental measurement approach and calculate each user's sentiment on things/products. Secondly, we have a tendency to not only think about a user's own sentimental attributes however additionally take interpersonal sentimental influence into thought. Then, we contemplate product reputation, that can be inferred by the sentimental distributions of a user set that replicate customers' comprehensive evaluation. At last, we tend to fuse three factors-user sentiment similarity, interpersonal sentimental influence, and item's name similarity-into our recommender system to create an accurate rating prediction. We tend to conduct a performance evaluation of the 3 sentimental factors on a real-world dataset collected from Yelp. Our experimental results show the sentiment will well characterize user preferences, which helps to enhance the recommendation performance.