Intelligent personalized recommender systems with trust assessment

Abstract

In recent years internet grows at a speedy rate Parallely heterogeneous information which has various types of data accumulates in cyber space Therefore the end users find very difficult to locate the relevant information satisfying their interests As a result recommendation systems appeared to help users is this task Recommendation system achieves widespread success in various domains such as E commerce social networking and advertisements The Implementation sequential approach in recommendation algorithm has large performance issues for a large dataset The performance issues have been addressed by implementing a novel algorithm to get recommendations by using efficient framework ensemble with an item based similarity collaborative filtering technique Recommendation system is a technique which works purely based on the user preference so it can provide an accurate prediction when enough data is provided On the other hand the widespread usage of recommender system has revealed some challenges such as data sparsity and data scalability with the mushrooming growth of items and users An item based collaborative filtering has been proposed to improve the execution time and accuracy of the prediction problem by applying similarity measure in a recommendation system It demonstrates that the proposed approaches can achieve better performance and execution time for the recommendation system in terms of existing challenges according to evaluation metrics using Mean Absolute Error MAE and trust worthiness newline