Comparing the accuracy of support vector machine and the Naïve Bayes algorithm : Twitter sentiment analysis during the COVID-19 pandemic

Almansour, Aseel

doi:10.47974/JSMS-1194

Open Access Research Article

Comparing the accuracy of support vector machine and the Naïve Bayes algorithm : Twitter sentiment analysis during the COVID-19 pandemic

^*Aseel AlmansourCorresponding authoraseel_almansour@yahoo.comDepartment of Statistics, Science CollegeKing Abdulaziz UniversityJeddah, P. O. Box. 80200, 21589, Saudi ArabiaView full profile →

* Corresponding author · click or hover a name for details

pp. 943–958Vol. 27Issue 5July 2024DOI: 10.47974/JSMS-1194XML

Received:: 15 Nov 2022
Published Online:: 05 Aug 2024
Article type:: Research Article
Language:: EN
Article no.:: JSMS-1194
Pages:: 943–958

Abstract

The COVID-19 pandemic has imposed substantial challenges globally, and many have voiced their thoughts on social media about the unfolding crisis. This study performed a sentiment analysis (SA) on Twitter posts, to assess the general public’s perspective on the pandemic. Our primary objective was to compare the performance of two frequently used algorithms, support vector machine (SVM) and Naive Bayes classification (NBC), in performing SA, which is a novel direction. The results showed that the more sophisticated SVM algorithm obtained 81% accuracy compared with the 74% achieved by NBC. Consequently, SVM may be used to assess and acquire improved public SA from Twitter.

Keywords

Sentiment analysis Naive bayes classification Support vector machine Machine learning COVID-19

Subject Classifications

62-08 Computational methods for problems pertaining to statistics

References

[1] Bhatnagar, V., Poonia, R. C., Nagar, P., Kumar, S., Singh, V., Raja, L., & Dass, P. Descriptive analysis of COVID-19 patients in the context of India. Journal of Interdisciplinary Mathematics (2020), DOI: 10.1080/09720502.2020.1761635. https://doi.org/10.1080/09720502.2020.1761635.
[2] Munjal, P., Narula, M., Kumar, S., & Banati, H. Twitter sentiments based suggestive framework to predict trends. Journal of Statistics and Management Systems, 21(4), 685-693 (2018). DOI: 10.1080/09720510.2018.1475079. https://doi.org/10.1080/09720510.2018.1475079.
[3] Singh, D. K., & Dass, P. On a functional equation related to some entropies in information theory. Journal of Discrete Mathematical Sciences and Cryptography, 21(3), 713-726 (2018). DOI: 10.1080/09720529.2018.1445809. https://doi.org/10.1080/09720529.2018.1445809.
[4] World Health Organization. (2020). Coronavirus disease (COVID-19) pandemic. https://www.who.int/emergencies/diseases/novel-coronavirus-2019.
[5] Ben, I. (2020). A simple statistical analysis for the spread of COVID-19. https://www.timesofisrael.com/israeli-mathematician-predicts-coronavirus-peaking-soon/.
[6] Mousavizadeh, L., & Ghasemi, S. Genotype and phenotype of COVID-19: Their roles in pathogenesis. Journal of Microbiology, Immunology, and Infection, 53(3), 409-412 (2020). DOI: 10.1016/j.jmii.2020.03.022. https://doi.org/10.1016/j.jmii.2020.03.022.
[7] Agarwal A., Xie B., Vovsha I., Rambow O., & Passonneau R. J. Sentiment analysis of Twitter data Proceedings of the workshop on language in social media, p 30–38 (2011).
[8] Bennett K. P., & Blue J. A. IEEE International Joint Conference on Neural Networks. A support vector machine approach to decision trees, 3, 2396–2401. (Cat No. 98CH36227) (1998).
[9] Borah P., Deb P. K., Deka S., Venugopala K. N., Singh V., Mailavaram R. P., Kalia K., & Tekade R. K. (2021). Current scenario and future prospect in the management of COVID-19. Current Medicinal Chemistry 27 (September), 28(2), 284–307. https://doi.org/10.2174/0929867327666200908113642
[10] Chen S., Webb G. I., Liu L., & Ma X. (2020). A novel selective naïve Bayes algorithm. Knowledge-Based Systems, 192, 105361. https://doi.org/10.1016/j.knosys.2019.105361
[11] Davidson H. (2020). First Covid-19 case happened in November. China government records show–report. The Guardian, 13.
[12] Dey S., Wasif S., Tonmoy D. S., Sultana S., Sarkar J., & Dey M. A comparative study of support vector machine and Naive Bayes classifier for sentiment analysis on Amazon product reviews. IEEE International Conference on Contemporary Computing and Applications (IC3A), p 217–220 (2020).
[13] Zhang K., Cheng Y., Liao W. K., & Choudhary A. Mining millions of reviews: A technique to rank products based on importance of reviews Proceedings of the 13th international conference on electronics and communications, p 1–8 (2011).
[14] Khorsheed M. S., & Al-Thubaity A. O. Comparative evaluation of text classification techniques using a large diverse Arabic dataset. Language Resources and Evaluation, 47(2), 513–538 (2013). https://doi.org/10.1007/s10579-013-9221-8
[15] Giachanou A., & Crestani F. Like it or not: A survey of Twitter sentiment analysis methods. ACM Computing Surveys, 49(2), 1–41 (2016). https://doi.org/10.1145/2938640.
[16] Go A., Huang L., & Bhayani R. Twitter sentiment analysis. Entropy, 17, 252 (2009).
[17] Jakkula V. Tutorial on support vector machine (SVM). School of EECS, 37. Washington State University, (p 2.5) : 3 (2006).

Views: 176Downloads: 75Citations: 0