A study on personalized recommender system using social media

• Main Author: Aishnivya, Balamurugan
• Format: Final Year Project / Dissertation / Thesis
• Published: 2020
• Subjects: T Technology (General); TA Engineering (General). Civil engineering (General)
• Online Access: http://eprints.utar.edu.my/3809/1/17ACB06183_FYP.pdf; http://eprints.utar.edu.my/3809/

Recommender system means to give user direction identified with the helpful services dependent on their personalised service suggestion, behaviour or neighbor's inclinations. With the popularity of social network, numerous clients like to share their perspectives via social networking media, for example, rating, sites, tweets and so on, which prescribes clients interest item. Personalised recommender framework gives a suitable recommendation like shopping, scheduling, hotels, tags, motion pictures and so on, which delivers enormous information on the web. This outcomes in the issue of data over-burden. To get over this issue, Personalized Recommendation System have been prosperously utilized. This paper studies personalised recommender system using social media. The problem that is discussed the research study is information overload which is a hot topic of current world. Objectives to condense the data , analyze the users’ behaviour before recommending items is much important. In the research study Naive Bayes Theorem classifier , k-Nearest Neighbor Classifier and Support Vector Machine classifier is used. These machine learning algorithm processes the data set obtained. The evaluation on these algorithm is done to evaluate accuracy of the algorithm. Therefore is also well stated the recommendation has a great play is almost everything. This assistance from our natural components gives us a basic technique to find the best choice without having a great deal of effort to isolating through the different choices available in the market. At this moment advancement, the Recommendation system is an application that isolated altered information and gives the best way to deal with understand a user's taste and to propose reasonable things to them by considering the models among their inclinations and research studies of various things. The research project has well stated the accuracy and determined performance evaluation metrics of the Twitter data set after going through the process of preprocessing and sentiment analyzing.