Author(s): S. Kalaimani1 , Dr. R.Mala2
There is several user recommendation systems are recently available but they are satisfying the user recommendation with the disturbance of unwanted data. So there is need for filtering the required data for user conformability. In this recommendation system, since they are growing up for various user needs. The user required information with unwanted data. The raw data transfer for information processing to classifying into separated groups, filtering the unwanted data, finally extracted the actual data as user recommended to make better decision to a particular standard. In earlier, the scientific community is concerned to increase the accuracy of different classification methods, and major achievements have been made so far. Cloud computing has become a feasible mainstream solution for data processing, storage and distribution. It confirms on demand, scalable, compute and storage capacity. So rich amount of data in cloud database or any other cloud file systems, for that Naïve Bayes and support vector machine (SVM) classification algorithms are used to discover knowledge. Recommender Systems apply for the machine learning and data mining techniques for filtering unseen information and can predict whether a user needed the given resource or not. Collaborative filtering recommender systems recommend items by identifying other users with same requirements and use their opinions for recommendation; such as content-based recommender systems recommend items based on the user needed information of the items. These kinds of various systems suffer from scalability, data parity, over specialization, and cold-start problems resulting in poor quality recommendations and lowest coverage. Hybrid recommended systems combine the individual systems to avoid certain limitations of these systems.