An efficient framework for the similarity prediction with query recommendation in E‐learning system

Abstract

Summary A novel model of data similarity estimation and clustering method is proposed in this article to retrieve the relevant data with the best matching in big data processing. An advanced model of graph distance pattern (GDP) method with lexical subgroup (LS) system is used to estimate the similarity between the query data and the entire database. With the help of neural network, the relevancy of feature attributes in the database are predicted and matching index is sorted to provide the recommended data for given query data. This was achieved by using the correlated sim‐neural network (CSNN). This is an enhanced model of neural network technology to find the relevancy based on the correlation factor of feature set. The training process of CSNN classifier is carried by estimating the correlation factor of the attributes of dataset. These are forms as the clusters and paged with proper indexing based on the LS parameter of similarity metric. The results obtained by the proposed system for recall, precision, accuracy, error rate, F‐measure, kappa coefficient, specificity, and MCC are 0.98, 0.98, 0.97, 0.03, 0.99, 0.991, 0.986, and 0.984, respectively.