Machine learning techniques for sentiment analysis

Abstract

Sentiment analysis stands out as one of the most dynamic and pivotal areas in the field of natural language processing. In this work, a range of machine learning strategies has been proposed, applied, and benchmarked for sentiment analysis, with a specific focus on supervised machine learning techniques. Various algorithms have been considered and applied to texts extracted from Twitter. Furthermore, the results are compared with works that applied unsupervised machine learning techniques to the same dataset.