Author : Dr. (Mrs.) N. F. Shaikh and Anindita Ghosh

Tourism has become popular across the world in important sectors making it a significant tool in improvement of a country’s economy. Nowadays with the boom of the internet services and webpages tourists want to be well prepared and well aware of the place they are planning to visit way ahead of their actual visit. This has become expected and possible because of the availability of large number of opinions and reviews about anything and everything. Similarly tourist reviews are information sources for tourists to know about tourist places beforehand. Users express their views and opinions regarding historical places and services or availability of resources. These opinions are subjective information which represents users’ sentiments, feelings or appraisal regarding the same. Unfortunately some reviews are either irrelevant or not from a true source. Thus they often become noisy data. Aspect-Based Sentiment classification has shown promise in suppressing such noise. When the aggregated data about the tourist places is presented in the right way, analyzed and classified by the correct aspect-based algorithm, it could be translated into meaningful information for making vital decisions by tourism enthusiasts. However, not much research has been done on automatic aspect identification, and identification of implicit, infrequent and co-referential aspects, resulting in misclassifications. In this paper we present a framework that employs different aspect-based algorithms to identify aspects and classify sentiments and opinions inherent in tourist reviews with high accuracy. The framework has been implemented as a mobile application that helps tourists find the best historical places of interest, and performance has been evaluated by conducting experiments on real-world datasets.

Keywords: Decision tree, classification, Clustering, Machine Learning, Aspect-based Sentiment Analysis, Opinion mining, products and services, travel, accommodations, Consumer Reviews, food, hotel rooms, Data mining, tourism etc.