Machine Learning Algorithms for Sentiment Analysis
Madhuri Sharma, Rajeev Mathur, Ganpat Joshi, Damodaran B, Komal Malsa, Pradeep Kumar · 2024
Sentiment analysis, a subset of natural language processing, focuses on identifying the emotional tone within text computationally. This study reviews machine learning algorithms used in sentiment analysis, assessing their effectiveness in interpreting sentiment from textual data. Algorithms such as Support Vector Machines (SVM), Naive Bayes, and Neural Networks are evaluated for their ability to classify sentiment polarity, from positive to negative. The research highlights the influence of feature selection techniques like n-grams and word embeddings on model performance. Additionally, challenges such as sarcasm detection, context dependency, and imbalanced data are explored. These obstacles complicate sentiment analysis but provide opportunities for further advancements. The findings offer valuable insights into the capabilities and limitations of various machine learning methods, supporting the development of more accurate and reliable sentiment analysis systems for diverse real-world applications across industries.