SCRF: Strategic Course Recommendation Framework

The need for course recommendation in the specialised domain is salient and therefore, this paper proposes a Strategic Course Recommendation Framework (SCRF) for recommending courses pertaining to finance and economics as a prospective domain. This paper encompasses a semantic-oriented learning and reasoning model for achieving accurate course recommendations in the era of Web 3.0. The model uses a knowledge stack comprising eBooks, Blogs, and Web pages relating to finance and economic domains. This knowledge stack is actually used to enhance the auxiliary knowledge between the data-enriched terms and the query-obtained terminologies. The semantic indexing and bag of words add to the auxiliary knowledge of the query to contextualize the query words that are pre-processed. Semantics-oriented learning is achieved through a strong deep learning classifier employed to classify the financial course dataset. The usage of Second Order Co-occurrence Pointwise Mutual Information (SOC-PMI), Normalized Pointwise Mutual Information (NPMI), and Pianka Index ensures strong relevance computation mechanisms at three distinct stages in the model to allow the convergence of knowledge with all of the data set enriched terms and the query enriched terms in order to facilitate the best-in-class foundations of courses in a highly cohesive web 3.0 environment. The proposed model, which is the best-in-class model when compared to all the other strategic models, has a total precision of 94.45%, a recall of 96.81%, and a False Detection Rate (FDR) of 0.06.