Health recommender system and its applicability with MapReduce framework

Abstract

There has been high influx of data over the Web in the past few decades. In order to convert this data into meaningful information, data analytics platforms such as big data analytics have emerged. Recommendation system (RS) is one such analytic system which uses predictive analysis to present and target the useful information to the users in order to help them arriving at desired conclusions. RS saves the time of most of the Internet users by reducing the number of actions needed to get to the required information which helps in effective decision making. RS has been widely used in multiple domains such as e-commerce, social networks to present recommended options to users in their current context. Recommended options thereby presented are based on the action taken by other users in similar context. RS is equally useful in healthcare services. In fact, they are a complementary tool to healthcare system which allows for effective decision making in healthcare services. RS uses either collaborative filtering approach, content filtering approach, or a blend of them (hybrid approach). This paper discusses the overview of what recommender systems are, how they are built, and its classifications. It also elaborates health recommender system (HRS) and gives a clear picture of how MapReduce Framework and Hadoop technology will help in improving the scalability and efficiency of HRS by stating illustrations.