A Cloud Based Personalised Recommendation System for movies and books

The cloud based personalized recommendation system , takes the current books, songs, movies present in your account by the ratings given to them and recommends new books , songs or movies based on the user interests.By using the techniques like content based and collaborative based, deep learning and matrix factorization techniques.

The main objective is to get all recommendation about different books, movies, songs in one place,to be more accurate ,giving ratings to them and have a faster response time by using various latest algorithms like deep learning and matrix factorzation.

The Existing system uses conventional data mining algorithms and are most specialized to cater a single need like movies alone or songs alone. And the main module is Recommendar system, where the user can view books, movies of different categories and can rate them as per his/her likings. Collaborative based and content based filtering techniques are used.

Recommendation system focuses on one domain alone . A user will have a good experience if it extends to more than one domain ,and accurate recommendations of books, songs, movies based on the user based on their interests.

Proposed system uses collaborative filter and deep learning algorithm and is not catered to a specialized category. Rather it focused on user and all his interest in different domain including songs, movies, books etc

The 4 recommendation systems that can be used. Here they are, in respective order of presentation:

Content-Based Filtering
Memory-Based Collaborative Filtering
Model-Based Collaborative Filtering
Deep Learning / Neural Network

Pearson correlation, Clustering algorithms

Other algorithms- Many algorithms — and an even larger set of variations of those algorithms Bayesian Belief Nets: which can be visualized as a directed acyclic graph, with arcs representing the associated probabilities among the variables.

Markov chains - which take a similar approach to Bayesian Belief Nets but treat the recommendation problem as sequential optimization instead of simply prediction.

Rocchio classification (developed with the Vector Space Model), which exploits feedback of the item relevance to improve recommendation accuracy.

We first started off with context based model, then proceeeded with model and memory based collaborative method finally we performed deep learning method. The accuracy of deep learning method was the highest.

Recommender systems are an extremely potent tool utilized to assist the selection process easier for users. The implemented recommendation engine is a competent system to recommend Books for e-users. This recommender system will definitely be a great web application implemented in Java language. Such type of web application will be proved beneficial for today‟s high demanding online purchasing web sites. This hybrid recommender system is more accurate and efficient as it combines the features of various recommendation techniques.The recommendation engine will reduce the overhead associated with making the best choices of books among the plenty.The future work can be focussed on improving the speed of the algorithm.