Publication: Sentiment Analysis on Mixed-Language Social Media Post
No Thumbnail Available
Date
2023
Authors
Mohamed Salleh F.H.
Gorment N.Z.
Mohd Ridza M.H.
Journal Title
Journal ISSN
Volume Title
Publisher
Institute of Electrical and Electronics Engineers Inc.
Abstract
Sentiment analysis is a powerful tool that can be used by businesses and organizations to gather valuable data about public opinion towards a brand, product, topic, event, and much more. However, most Malaysians post on social media using a mix of English and Malay words, also known as Manglish, which are not catered to by existing sentiment analysis models. Malaysian-centric companies that are interested to analyze the Malaysian posts would have to do so manually, which is costly and time-consuming. Motivated by this issue, this paper aims to propose a method of performing sentiment analysis on posts using the power of machine learning. Several machine learning algorithms were identified and trained to classify a Manglish post as either positive or negative. Steps are also taken to ensure the reliability of the model and to improve it after the first training experiment. We found that this method is successful in producing a model that can predict social media post sentiment with reliable and acceptable accuracy. � 2023 IEEE.
Description
Keywords
machine learning , mixed-language , sentiment analysis , social media post , Learning algorithms , Machine learning , Social aspects , Social networking (online) , Analysis models , Machine-learning , Malaysians , Mixed-language , Power , Public opinions , Sentiment analysis , Social media , Social medium post , Topic events , Sentiment analysis