An Enhancement of Support Vector Machine in Context of Sentiment Analysis Applied in Scraped Data from Tripadvisor Hotel Reviews

  • Lambert T. Dela Cruz College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines
  • Marjorie Jasmine C. Racelis College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines
  • Vivien C. Agustin College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines
  • Jamillah S. Guialil College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines
  • Gabriel R. Hill Senior Software Engineer, Accenture, Philippines
  • Leisyl M. Mahusay College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines
  • Jonathan C. Morano College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines

Abstract

Purpose – The purpose of this study is to improve the efficiency and accuracy of sentiment analysis in the context of hotel reviews, thereby contributing to the advancement of machine learning and natural language processing fields.

Method – The study employs an Enhanced SVM algorithm, incorporating SMO, Random Search, and SMOTE, to address issues of long training time, hyperparameter optimization, and imbalanced data.

Results – The Enhanced SVM outperforms the Traditional SVM, with a 13.48% increase in accuracy, a 100.42% reduction in training time, and improvements of 11.5% and 8.5% in Precision and F1-Score, respectively.

Conclusion – The study successfully enhances the SVM algorithm, providing a more effective tool for sentiment analysis in the context of hotel reviews, with significant improvements in performance metrics.

Recommendations – Future researchers should explore advanced optimization methods for hyperparameter tuning, use additional linguistic features like semantic analysis and context-aware embeddings, and incorporate sarcasm detection. Furthermore, consider deep learning models and ensemble approaches, combining SVM with other algorithms. Lastly, advocating for real-time sentiment analysis is suggested for immediate customer feedback insights.

Research Implications – The study offers valuable insights into the application of machine learning techniques in sentiment analysis, particularly in the tourism industry.

Practical Implications – The Enhanced SVM model can be used by platforms like TripAdvisor to provide more accurate sentiment analysis of hotel reviews, aiding tourists in their decision-making process.

Social Implications – Improved sentiment analysis can enhance the overall travel experience, leading to more satisfying and informed travel decisions.

Author Biographies

Lambert T. Dela Cruz, College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines

Lambert T. Dela Cruz, a Computer Science student at Pamantasan ng Lungsod ng Maynila, has already distinguished himself as a consistent academic achiever and former PLM Computer Science Society President. His leadership extended to his role as Cloud Development Lead at Google Developers Student Clubs PLM. Professionally, Lambert has made significant strides as a Google Cloud Engineer at Tenet Global Business Center, holding certifications as an AWS Certified Cloud Practitioner and AWS Certified Solutions Architect Associate. He also possesses a Hashicorp Terraform Associate certification, underscoring his expertise in infrastructure automation.  

Marjorie Jasmine C. Racelis, College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines

Marjorie Jasmine C. Racelis, a Computer Science student at Pamantasan ng Lungsod ng Maynila, possesses an excellent academic record. Her professional experience includes working as an IT Project Manager at XScribe Solutions, Inc., where she displayed exceptional project management and quality assurance skills. Marjorie has also completed internships at DFI Retail Group, Creative Interlace Technologies, and Dashlabs.ai, where she strengthened her technical and leadership skills. As a core member of the Google Developer Student Club PLM and a former CSS Vice President of the PLM Computer Science Society, she has demonstrated a strong dedication to community and organizational growth.  

Vivien C. Agustin, College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines

Vivien C. Agustin serves as a Thesis Adviser at the College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines. With a deep understanding of information technology and systems, Vivien guides students through the rigorous process of thesis development, ensuring that their research is both innovative and impactful.

Jamillah S. Guialil, College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines

Jamillah S. Guialil is a Thesis Panelist at the College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines. As an expert in information systems, Jamillah plays a critical role in evaluating and providing feedback on student theses, ensuring they meet academic and industry standards. Her insights help students refine their research and strengthen their projects.

Gabriel R. Hill, Senior Software Engineer, Accenture, Philippines

Gabriel R. Hill serves as a Thesis Panelist at the College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines, in addition to his role as a Senior Software Engineer at Accenture Philippines. Gabriel brings his extensive experience in software development and engineering to the evaluation of student theses, offering a practical perspective on the application of their research. 

Leisyl M. Mahusay, College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines

Leisyl M. Mahusay is the Thesis Coordinator at the College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines. In her role, Leisyl oversees the thesis process, ensuring that all aspects, from proposal to defense, are conducted smoothly and efficiently. 

Jonathan C. Morano, College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines

Jonathan C. Morano is a Thesis Coordinator at the College of Information Systems and Technology Management, Pamantasan ng Lungsod ng Maynila, Philippines. In this role, Jonathan is responsible for managing the thesis process, ensuring that students receive the guidance and resources they need to complete their research successfully.

Published
2024-08-26
How to Cite
DELA CRUZ, Lambert T. et al. An Enhancement of Support Vector Machine in Context of Sentiment Analysis Applied in Scraped Data from Tripadvisor Hotel Reviews. International Journal of Computing Sciences Research, [S.l.], v. 8, p. 3235-3251, aug. 2024. ISSN 2546-115X. Available at: <//stepacademic.net/ijcsr/article/view/616>. Date accessed: 22 dec. 2024.
Citation Formats
Issue
Vol 8 (2024): Volume 8
Section
Articles

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.

Most read articles by the same author(s)