Implementation of a map-reduce based context-aware recommendation engine for social music events
|Title||Implementation of a map-reduce based context-aware recommendation engine for social music events|
|Journal||International Journal On Advances in Intelligent Systems|
In our modern ubiquitously connected world the amount of ever available product and service information within our daily lives is exploding. Powerful client devices, such as smartphones and tablets allow the users to get access to an unlimited amount of information on every product or service available. As the amount of available information on products by far exceeds the users time to examine and filter detailed pieces of information in every situation, we expect that client-centric and context-aware information filtering is one of the thriving topics within the next years. A popular approach is to combine context-awareness with traditional recommendation engines in order to evaluate the relevance of a large amount of items for a given user situation. The goal is to proactively evaluate the situation of a user in order to automatically propose relevant products. Within this work we describe a general approach and the implementation of a software framework that combines traditional recommendation methods with a variable number of context dimensions, such as location or social context. The main contribution of this work is to show how to use a MapReduce programming model for aggregating the necessary information for calculating fast context-aware recommendations as well as how to overcome a typical cold start problem. The use-case at the end of this work evaluates the practical benefit of our general framework to introduce a client-centric, MapReducebased recommendation engine for real-time recommending music events and festivals.