A framework which is complaint to the Web 3.0, in which learning-inferencing through models for deep learning and machine learning along with Semantics Intelligence techniques has been proposed. This paper suggests a data set enrichment strategy for the key words extracted from the data set through structured topic modelling and the addition of terms from Wikidata and knowledge repositories. Subsequently the query is also enriched using meta data which is generated from the dataset which is further classified using transformers and formalized into a knowledge graph which uses Google Knowledge Graph API to be enhanced further. So the strategic encompassment and outgrowing of the dataset through topic modelling and auxiliary knowledge store repositories and knowledge graph through the meta data which ensures strong degree of knowledge density into the model with semantic relatedness computation using normalized compression distance, normalized information distance and Morisita’s overlap index along with Pearsons correlation coefficient helps in semantics rigidness measure based reasoning and all baseline models are outperformed by the suggested framework. and achieves a overall Precision of 93.45% and F- measure of 189.30% and having a lowest FDR of 0.07, which model is the best in its class for microblog tagging in a highly dense Semantic Web environment.