Location Aspect Based Sentiment Analyzer for Hotel Recommender System

  • Alia Karim Abdul Hassan Department of Computer science, University of Technology, Baghdad, Baghdad, Iraq
  • Ahmed Bahaa Aldeen Abdulwahhab Department of Informatics, Middle Technical University, Baghdad, Baghdad, Iraq
Keywords: sentiment analysis, recommender systems, location-based, natural language processing, ABSA

Abstract

Recently personal recommender system has spread fast, because of its role in helping users to make their decision. Location-based recommender systems are one of these systems. These systems are working by sensing the location of the person and suggest the best services to him in his area. Unfortunately, these systems that depend on explicit user rating suffering from cold start and sparsity problems. The proposed system depends on the current user position to recommend a hotel to him, and on reviews analysis. The hybrid sentiment analyzer consists of supervised sentiment analyzer and the second stage is lexicon sentiment analyzer. This system has a contribute over the sentiment analyzer by extracting the aspects that users have been mentioned in their reviews like (cleanness, service, etc.) by using accurate parsing system built on latent semantic analysis results. The accuracy measurements of the proposed sentiment analyzer were perfect.

Published
2019-01-28
How to Cite
Abdul Hassan, A. K., & Abdulwahhab, A. B. A. (2019). Location Aspect Based Sentiment Analyzer for Hotel Recommender System. Iraqi Journal of Science, 60(1), 143-156. Retrieved from https://ijs.uobaghdad.edu.iq/index.php/eijs/article/view/465
Section
Computer Science