Hyperparameter Optimization of Machine Learning Models Using Grid Search for Twitter Sentiment Analysis

Abstract

Twitter has emerged as a significant social media platform and has garnered significant interest from sentiment analysis researchers. Text mining is an active subfield that includes Twitter sentiment analysis (TSA) research. TSA is the term used to describe the utilization of algorithms to analyze the subjective nature of Twitter data, which includes its sentiments and opinions. The extraction of inferences from user interactions is facilitated by machine learning (ML) approaches. A wide range of machine learning methodologies are employed to analyze emotions. This research compares four supervised machine learning techniques with the term frequency-inverse document frequency (TF-IDF) method and hyperparameter tuning with grid search for the Dell tweets dataset. This classification technique will demonstrate how sentiment analysis is performed using performance metrics.