OntoDynS: expediting personalization and diversification in semantic search by facilitating cognitive human interaction through ontology bagging and dynamic ontology alignment

To reduce the cognitive gap between human cognition and semantic search engines, there is a need for the incorporation of crowdsourced Ontologies and closed-world reasoning for recommending contents from the Web 3.0. In this paper, an OntoDynS framework has been proposed to tackle the ambiguity and the serendipity problem in recommending queries and URLs from the Semantic Web. A novel scheme called Ontology Bagging based on lateral co-similarity and collective similarity has been proposed to encompass diversity in the search results. OntoDynS adapts the Second Order Co-occurrence Pointwise mutual information in a semantic environment and proposes a novel Blend of Similarity measure with scenario-based weighting for query recommendation for yielding initial recommendations. Personalization is achieved by capturing current user-clicks through Dynamic Ontology Alignment by the integration of agent-based semantic rules for yielding minimal mappings for diversified user-centric query recommendation which is succeeded by recommending web pages. The dynamic amalgamation of crowdsourced Ontologies and Open Linked Data facilitates human–computer interaction. OntoDynS yields an F-Measure of 94.70% and 95.22% for recommending queries and web pages respectively for the DMOZ dataset. Experiments on CACM datasets have furnished an F-Measure of 76.56% and 77.71% for recommending queries and web pages respectively.