Sentiment Analysis Technique and Neutrosophic Set Theory for Mining and Ranking Big Data From Online Reviews

Recently, a huge amount of online consumer reviews (OCRs) is being generated through social media, web contents, and microblogs. This scale big data cannot be handled by traditional methods. Sentiment analysis (SA) or opinion mining emerging as powerful efficient tool in analytics improving decision making. research paper introduces novel method that integrates neutrosophic set (NS) theory into the SA technique multi-attribute making (MADM) to rank different products based on numerous reviews. The consists two parts: Determining sentiment scores ranking alternative via NS theory. In first part, concerning multiple features are crawled pre-processed. A neutral lexicon 228 words phrases compiled Valence Aware Dictionary sEntiment Reasoner (VADER) for reasoning adapted handle data. lexicon, well VADER, utilized build adaptation called Neutro-VADER. Neutro-VADER assigns positive, neutral, negative each review product feature. this stage, idea point out truth, indeterminacy, falsity memberships degrees number. overall performance feature number measured. second alternatives evaluated simplified weighted averaging (SNNWA) operator cosine similarity measure case study with real datasets (Twitter datasets) provided illustrate application proposed method. results show good handling stage stage. findings can deal successfully all types uncertainties including indeterminacy comparable VADER other great consistency while using methods such PROMETHEE II, TOPSIS, TODIM