Detecting and Eliminating Negative Chaos Creating Tweets Using Crisis-Causing Tweet Elimination (CCTE) Algorithm

Opinion mining is a subfield within the realm of natural language processing (NLP), which is dedicated to the systematic analysis of human sentiment contained within textual data. In light of the pervasive proliferation of e-commerce platforms, microblogging sites, and social media networks, OM within the context of online social media platforms has garnered substantial attention from a multitude of scholarly investigators. In the proposed methodology, the attributed realization that reviews, tweets, and blogs originating from these social media platforms serve as significant reservoirs of data with the potential to greatly augment the decision-making process. The textual data harvested from these sources, encompassing reviews, tweets, or blogs, undergo a classification process that segregates them into three distinct class labels: negative, neutral, and positive from twitter. This categorization serves the purpose of facilitating the systematic analysis and extraction of pertinent information from the provided dataset. From the extracted words, classifier analyzes and identifies crisis-causing words from the tweets, compared with training datasets, and categorizes them into negative crisis-causing words using crisis-causing tweet elimination (CCTE) algorithm. Those messages containing these words will be blocked and deleted without displaying in twitter. This avoids major crisis caused in group of people agitated or provoked by chaos creating tweet texts.