Web Interaction Mining Using Adaptive Feature Selection Basedpenta Layered Artificial Neural Network (AFS-Plan) Classifier

Abstract

Predicting the objective of internet users contains different applications in the areas such as e-commerce, entertainment in online, and several internet-based applications. The integral section of classifying the internet queries based on accessible features such as contextual information, keywords and their semantic relationships. This research article aims in proposing Adaptive Feature Selection based Penta Layered Artificial Neural Network Classifier for web interaction mining. Around 31 participants are chosen and given topics to search web contents. Parameters such as precision, recall and F1 score are taken for comparing the proposed AFS-PLANN classifier with the ANN and PLANN. Results proved that the proposed classifier attains better performance than that of the conventional ANN and PLANN.