Perbandingan Kinerja Machine Learning dan Deep Learning untuk Analisis Sentimen Fufufafa

Abstract

This study examines the performance comparison of various machine learning and deep learning models in sentiment analysis of the fufufafa phenomenon on Twitter. The models tested include Naive Bayes, Random Forest, Support Vector Machine, Logistic Regression, Decision Tree, and Long Short-Term Memory (LSTM). The study aims to evaluate the effectiveness of these models in classifying sentiment into positive, negative, and neutral categories. Experimental results show that the Decision Tree model achieved the highest accuracy of 96%, followed by LSTM with 95%. Other models, such as Random Forest, Support Vector Machine, and Logistic Regression, achieved an accuracy of 94%, while Naive Bayes recorded the lowest accuracy at 81%, primarily due to its limitations in handling more complex sentiments. The LSTM model proved to be superior in capturing temporal contexts and word relationships, delivering more accurate predictions despite requiring greater computational resources. The findings of this study affirm that deep learning models, particularly LSTM, are more effective in analyzing sentiment in dynamic and complex data, such as that found on Twitter, compared to machine learning models. These results provide valuable insights for developing more efficient and accurate sentiment analysis methodologies in the future.