Machine learning-based sentiment analysis for tweets saudi tourism

SARAH M ALRASHIDI; FATMH N ALANAZI; HANAN A ALBALAWI; OHOOD M ALBALAWI; AWAD M AWADELKARIM

Author(s)

SARAH M ALRASHIDI Faculty of Computers and Information Technology, University of Tabuk,Tabuk 47731,Saudi Arabia

FATMH N ALANAZI Faculty of Computers and Information Technology, University of Tabuk,Tabuk 47731,Saudi Arabia

HANAN A ALBALAWI Faculty of Computers and Information Technology, University of Tabuk,Tabuk 47731,Saudi Arabia

OHOOD M ALBALAWI Faculty of Computers and Information Technology, University of Tabuk,Tabuk 47731,Saudi Arabia

AWAD M AWADELKARIM Faculty of Computers and Information Technology, University of Tabuk,Tabuk 47731,Saudi Arabia

Abstract

The growth of tourism in Saudi Arabia is a superior aspect of the recent economic success and realization of the Kingdom of Saudi Arabia's Vision 2030. This places a premium on research in such field and elevates it to national priority research, thus, this research project contributes to such context and places a premium on sentiment analysis in the tourism industry, namely concentrating on tweets about Saudi tourism. Therefore, this paper intends to demonstrate machine learning-based sentiment analysis models for tweets on Saudi tourism. The research studies and analyzes tweets related to tourism collected about six touristic places in Saudi Arabia from twenty accounts and nineteen touristic places from 144 hashtags. Following the preprocessing and feature extraction stage, such tweets are labeled as positive or negative using various machine learning algorithms. Two base classifier models of Support Vector Machine (SVM) and Naïve Bayes (NB) are applied. Over and above, a vital and important contribution of this project is creating the First Dataset for Tweets Saudi Tourism (FDTST) in both Arabic and English languages collected from Twitter libraries. Utterly, numerous classification models are developed and evaluated based on their performance computation, and the experimental results show that the developed models have achieved righteous and reliable upshots. Finally, the developed predictive models aid to appoint and specify several valuable recommendations and insights for continuous improvements and sustainable growth in the Saudi tourism industry.

Publication Details

Page(s) 5096-5109

DOI DOI not available

Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 100, Issue: 16, Year: 2022

Request the full text from PASTIC

Keywords

sentiment analysis, machine learning, Saudi Tourism, Classification Models, Tweets Saudi Tourism