Aspect Based Sentiment Classification Using Machine Learning for Online Reviews

EasyChair Preprint no. 3051

Abstract

The tourism and travel sector is improving services using a large amount of data collected from different sources. The easy access to comments, evaluations and experiences of different tourists has made the planning of tourism rich and complex. Therefore, a big challenge faced by tourism sector is to use the gathered data for detecting tourist preferences. Unfortunately, some user’s comments are irrelevant and complex for understanding these becomes hard for recommendation. Aspect based sentiment classification methods have shown promise in overcome the noise. In existing not much work on aspect based sentiment with classification. This paper presents a framework of aspect based sentiment classification recommendation system that will not only identify the aspects very efficiently but can perform classification task with high accuracy using machine learning naıve Bayes and Decision Tree algorithms. The framework helps tourists find the best place, hotel and restaurant in a city, and performance has been evaluated by conducting experiments on Yelp and foursquare real-time datasets.

Keyphrases: aspect-based sentiment analysis, consumer reviews, machine learning, text mining

@Booklet{EasyChair:3051,
  author = {Pratiksha Nehe and A.N. Nawathe},
  title = {Aspect Based Sentiment Classification Using Machine Learning for Online Reviews},
  howpublished = {EasyChair Preprint no. 3051},

  year = {EasyChair, 2020}}