Building Personalized and Non Personalized Recommendation Systems

For a modern life style the contents of e-Commerce such as music, movies and electronics goods have become a necessity. There are many things which one needs in daily life which needs to be searched over the internet. There are many filtering techniques because of which the data is filtered and instead of getting access to every data available, we get relevant results. But, it becomes difficult to find results according to users' likes and users' preference. An approach which produces desirable results to solve such the problem is to develop "Recommender System." The Recommender System of an e-Commerce site selects and suggests the contents to meet user's preference automatically using data sets of previous users stored in database. There can be two types of recommendations viz. Personalized and Non-Personalized recommendations. Personalized recommendation takes into consideration users' previous history for rating and predicting items. On the other hand non-personalized recommendation systems recommend what is popular and relevant to all the users which can be a list of top-10 items for every new user. One of the most important techniques in the Recommender System is information filtering. The filtering techniques can be mainly classified into two categories viz. Collaborative Filtering and Content Based Filtering. Recommender system is a type of web intelligence technique that can make daily information filtering for users. This paper covers different techniques which can be used for creating personalized and non-personalized recommendations. This paper also explores the different packages of R i.e. Shiny which is used to create web applications and rmarkdown which is used to create dynamic documents.