Sentiment Analysis on Generation Z News Article using Support Vectore Machine (SVM) with Synthetic Minority Over-sampling Technique (SMOTE)

Abstract

The development of digital media has increased the volume of news articles discussing various issues, including those involving Generation Z. Understanding public perception of these news items can be achieved by applying a crucial approach, namely sentiment analysis. This study aims to classify sentiment in news articles about Generation Z using the Support Vector Machine (SVM) algorithm. The main challenge in sentiment analysis is data class imbalance, where the amount of positive and negative sentiment data is often unbalanced. Therefore, the Synthetic Minority Over-sampling Technique (SMOTE) is used to address this problem by balancing the class distribution before model training. The datasets used were collected from various online news portals and analyzed through text preprocessing, feature extraction using Bag of Word, and SVM model training. The evaluation results show that the application of SMOTE significantly improves the model's performance in classifying sentiment, with improvements in accuracy, precision, recall, and F1-score compared to the model without data imbalance handling. This study demonstrates that the combination of SVM and SMOTE is effective in conducting sentiment analysis on Generation Z news articles. The accuracy shows 84% with 83% precision and 76% recall.