Model Machine learning for sentiment analysis of the presence of electric cars in Indonesian
Amril mutoi Siregar, Sutan Faisal, Ahmad fauzi, Jamaludin Indra, Anis Fitri Nur Masruriyah, Adi Rizky Pratama

Department of Informatics Engineering, Faculty of Computer Science, University of Buana Perjuangan, Karawang, Indonesia

Abstract

Sentiment analysis, also known as opinion analysis or social sentiment analysis, is a well-established field of study. Within the automotive industry, great attention is being paid to the presence of electric cars as a viable solution to the pressing issue of greenhouse gas emissions. In order to gauge the level of acceptance and adoption of this technology, it is crucial to analyze the sentiments and opinions expressed by individuals towards electric cars. Various approaches can be employed for sentiment analysis, including rule-based techniques, statistical methods, and machine learning algorithms. The objective of this research endeavor is to conduct sentiment analysis on online publications and social media discussions pertaining to electric cars. Logistic Regression (LR), Support Vector Machine (SVM), and Random Forest (RF) are the specific methods employed in this study. The effectiveness of these methods is evaluated using accuracy measurements and Receiver Operating Characteristic (ROC) analysis. The accuracy outcomes attained by LR were 78.02%, SVM achieved 71.92%, and RF exhibited 82.35%. By virtue of the examination outcomes of multiple techniques utilized, there is an optimistic expectation that this can serve as the initial stride towards constructing sentiment applications for the existence of electric cars in the Indonesian context

Keywords: electric car- sentiment analysis- machine learning

Topic: Life Sciences