Social Network Data Analysis using Recommender System and Sentiment Analysis

Abstract

Social Network Data Analysis using Recommender System and Sentiment Analysis, offers a wide range of opportunity to work in the domains like policy building, decision support system, election prediction system, health system, reviewing system or predicting the future needs. It promotes the development of the application emphasis on the feeling of the people. Around three decades of research have produced many techniques for recommendation and Sentiment Analysis on various kinds of data. newlineRecommender system as the most successful application of information filtering help users to find items of their interest from huge dataset. To improve the performance of the Recommender System, Trusted users are identified and trustworthy value on relation among users to make more reliable accurate recommendation. The work has been done to find the trusted user and calculate the trusted score of each user and a threshold of 60 is considered as trusted user. Apriori algorithm of data mining is used for finding the frequent movie genre combination and their score value which would be helpful for new movie genre recommendation. To validate the above methodology different test cases containing fifty user each, the average accuracy achieved by different test cases is 88.67 %. newlineThe course of dimensionality is a damning factor for numerous potentially powerful machine learning techniques. Widely approved and otherwise elegant methodology used for many different tasks ranging from classification to function approximation, exhibit high computation complexity with respect to dimensionality. Rough Set Theory is a formal methodology that can be employed to reduce the dimensionality of datasets as a pre-processing step to train a learning system on the data. The performance of the proposed methodology increases by 30 percent over without using feature selection technique. newlineIn recent years the Sentiment Analysis has been widely used by researchers to recommend contents in accordance with human emotions, which are expressed through informal text