Volume 11 Issue 1 ( March )

Sentiment analysis, a subfield of natural language processing (NLP), has grown signif-icantly in importance and omplexity. This research introduces an innovative framework for handling binary clustered sentences, a prevalent hallenge in sentiment analysis. This approach groups sen-tences into positive or negative clusters and determines the sentiment of ach cluster based on the majority of sentences within it, enhancing the overall accuracy of sentiment analysis. Another ver-looked yet crucial aspect, the impact of negation and double negation on sentiment polarity, is also addressed. Current odels often fail to capture these linguistic nuances, hindering a complete under-standing of the true sentiment in the ext.The research also introduces the FFBC algorithm, specifi-cally designed to handle complex linguistic constructs like egations and double negations, often overlooked in current models. Validated on IMDb and Amazon Reviews Datasets, nd tested on a unique Farmers' Protest Twitter dataset, the framework shows enhanced performance across key met-rics ompared to leading techniques like BERT, LSTMs, VADER, and SVM. This improvement un-derscores the potential of dvanced sentiment analysis techniques in the digital era, offering signifi-cant insights into public sentiment during global vents. The study concludes by highlighting the implications of this research for various stakeholders and outlining future esearch directions.

Keywords: Binary-Clustered Sentences; Double Negation; Farmers Protest; FFBC Technique; Majority Voting System, Machine Learning; Sentiment Analysis.