Engineering College Decision Making Assistant

Abstract

People can voice their opinion which has allowed them to write a review and comment about how they feel about something. There can be millions of reviews on the internet for services which makes it difficult to track and understand the opinions of the customer.  An emerging area of research to extract the subjective information to track and understand opinions of the customer is by using Sentiment Analysis. Accessible and plentiful data is been provided by the reviews for relatively easing the analysis for a wide range of applications.[1] This system seeks application and extension of the current work in the field sentiment analysis on data retrieved about college reviews from other websites using web mining technique.[2] A given review can be tagged as positive or negative by using Naive Bayes and decision list classifiers. Features such as bag-of-words and bigrams are compared to one another in their effectiveness in correctly tagging reviews. Recent studies analysed these reviews and found that it includes information useful for colleges, such as user requirements, ideas for improvements, user sentiments about specific features and descriptions of experiences with these features. In this project, prediction of next year’s cut off on the basis of the last five years. College suggestions are also given based on percentage.