Study and analysis of various optimisation techniques for sentiment classification: a challenging overview

Sentiment classification aims to accurately rate a text's polarity toward a score or label that indicates whether the text's opinion is positive, negative or neutral. Short text reviews, in contrast to conventional expression, exhibit a high level of colloquialism, irregular grammatical structures and low sentence integrity. This survey's primary focus is on a variety of techniques used to classify sentiment based on online reviews. This study provides a technique-by-technique review of 50 research papers that focus on various techniques. The implementation software, performance and other factors are the foundation for the comprehensive analysis. The analysis is promoted in the survey based on the publication year, research technique, performance measures and achievement of the research methodologies towards sentiment classification. In the end, it identifies the techniques' problems and research gaps in a way that makes it clear why an effective method is needed to make efficient improvement possible.