Sentiment Analysis of Online Movies? Reviews Using Improved k-Nearest Neighbor Classifier

摘要

Others’ opinion can be decisive while choosing among various options; especially when those choices involve worthy resources like spending time and money buying products or services. A significant reason behind information gathering of opinions has always been there to figure out what other people think. Until a decade ago, the only sources of information were family, friends or acquaintances. But now, electronic word-of-mouth has become a flourishing frontier. Customers relying on their peers’ past reviews on e-commerce websites or social media have drawn a considerable interest to sentiment analysis due to the realization of its commercial and business benefits. In this paper, sentiment analysis of movie reviews has been carried out to ascertain the sentiment behind the review- whether positive or negative and an improved k- Nearest Neighbor (ImpkNN) classifier has been designed which uses the concept of attribute weighted-kNN and the weights associated are trained using cross validation. At last, the results of both Basic kNN and ImpKNN are evaluated using graphs.