Utilizing Random Forest Algorithm for Sentiment Prediction Based on Twitter Data
Iwan Setiawan, Agung Mulyo Widodo, Mosiur Rahaman, Tugiman Tugiman, Muhammad Abdullah Hadi, Nizirwan Anwar, Muhamad Bahrul Ulum, Erry Yudhya Mulyani, Nixon Erzed · 2022
Information sharing throughout the globe or universe has become a characteristic of social media.There has been a lot of research into the classification of sentiments.In this study, Twitter has been mined for unstructured GoFood Reviews data.It has been preprocessed to analyze the reviews' sentiment with polarity analysis, feature extraction with TF-IDF, and supervised learning with random forest.From June 1, 2022, to June 30, 2022, a total of 28763 tweets with the keyword GoFood were retrieved from Twitter.The data is processed by the Python programming language utilizing NLTK, Sastrawi for the Indonesian language, Textblob, TF-IDF, Random Forest Classification, and other algorithms.Twitter is a nearly limitless source for classifying text.This algorithm takes roughly five minutes to compute.