Predicting best answer using sentiment analysis in community question answering systems

While interests in seeking and sharing questions/ answers through the Community Question Answering (CQA) systems has been increased, predicting the best answer in such systems is one of the main challenges that we are going to tackle in this paper. Considering comments as one of the inputs in our model and extracting features using Natural Language Processing (NLP) and text mining techniques such as Sentiment Analysis (SA) on comments and spell checking for answers, are the main parts of this research. Moreover, we worked on English language websites. On the other hand, users' social behavior and their activities considered as informative features in this paper. As a result, by finding the best combination of different features the performance of our model shows improvement in comparison to the related previous works on "Stack Exchange" websites.