Social Media Analysis using Fuzzy Natural Language Processing with an extension of semantic queries

Information retrieval (IR) is a topic of continuing study, with the apparent goal being to find the most relevant data in massive repositories. The user's inquiry is essential at this stage. Finding crucial documents written in natural language has required extensive effort. Expertise retrieval relies on query extension to provide accurate data retrieval. This research presents a semantic query extension for social media analysis utilizing fuzzy natural language processing. This model computes the sentiment of social media posts using an innovative set of fuzzy rules informed by a wide range of lexicons and data sets. Using a combination of NLP and fuzzy logic, the proposed technique classifies posts according to whether they were written with joy, sadness, or no particular emotion. Researchers can use the results to zero in on the ideal social media jargon. Based on experimental results, it is clear that the suggested model can guarantee effective performance on semantic query expansion, outperforming state-of-the-art models in terms of classification accuracy (97.32 percent). Based on the studies mentioned above, we find that the success rate of enhancing fuzzy natural language is 85.7
^%
, the success rate for boosting precision on text search engines is 85.3%, the success rate of exploring records from databases is 83.6%, the success rate of investigating the lexicons of social media conversations is 82.5%, the success rate of developing the knowledge retrieval system is 91.84%, and the success rate of decreasing query extensions is 84.6%.