Sentiment-based search in digital libraries

Several researchers have developed tools for classifying/ clustering Web search results into different topic areas (such as sports, movies, travel, etc.), and to help users identify relevant results quickly in the area of interest. This study follows a similar approach, but is in the area of sentiment classification -automatically classifying on-line review documents according to the overall sentiment expressed in them. This paper presents a prototype system that has been developed to perform sentiment categorization of Web search results. It assists users to quickly focus on recommended (or non-recommended) information by classifying Web search results into four categories: positive, negative, neutral, and non-review documents, by using an automatic classifier based on a supervised machine learning algorithm, Support Vector Machine (SVM).