Indexing documents and Web pages is a vital task in the modern-day world as the documents and Web pages are increasing on the World Wide Web. In this paper, the RISL which is a framework for the metadata driven RDF centered model for indexing Web documents has been proposed. The model is based on Entity Enrichment using the Structural Topic Model. The Subject and Object pairs generated in the RDF are retained and aggregated with the upper ontologies which not only provision lateral term pair semantics but also serves as an active model to increase the knowledge density. Furthermore. The encompassment of Googles Knowledge Graph API for entity enrichment based on knowledge graph synthesis, andthe classification of dataset using the ensemble.Bagging model which inturn integrates Support Vector Machine and Random Forest classifiers facilitate in anchoring the right knowledge for generation of indexes. The semantic similarity computation using the Adaptive Pointwise Mutual Information (APMI) measure and the deviance computing criteria using the Heips Evenness Index and the Pearson Correlation Coefficient at varied instances with different levels of heterogeneity ensures strong relevance computation schemes. The RISL framework encompasses the Lion Optimization scheme for refining the feasible solution set to derive optimal solution set for indexing Web documents. The proposed RISL framework yields an overall precision of 95.12 with the lowest FDR of 0.05 which is the best in class model for generating indexing of Web documents

RDF, Indexing, Intelligent Semantics, Lion Optimization, Semantic Similarity