In today’s advanced and fast running world, there is an enormous availability of digital books on the internet. But getting a proper book suggestion is still a question mark. So, several machine learning algorithms are being developed for addressing this prevailing problem. Therefore, research is being conducted to improve the recommendation model to enhance the diversity in recommendation. In this paper, a Hybridised Knowledge Centric Book Recommendation model is proposed using the hybridization of SemantoSim measure, Kullback-Liebler divergence with optimization done by novel cultural algorithm. The model has incorporated structural topic modelling, named entity recognition to uncover the related documents from the internet. The static domain ontology is aligned with upper ontology using Lin similarity measure. Indexes are then matched with the ontological terms and crawled from google books metadata repositories. The proposed system HyKCBookRec also has proven better performing model than BFRE, CFJS, CBRF, PBRDL models with an overall precision and Recall of 96.72% and 98.87% respectively

Book Recommender, Integrative Semantics, Knowledge Centric, SemantoSim, Variational Learning