A Distributed System for Finding High Profit Areas over Big Taxi Trip Data with MognoDB and Spark
Fadhilah Kurnia Putri, Joonho Kwon · 2017
Due to the rapid development of internet of things (IoT) technology, traditional taxi cabs are connected through dispatchers and location systems. In major urban cities, modern taxis have equipped with GPS (Global positioning system) sensors which generate huge volumes of taxi trips. One of crucial solutions from analyzing taxi trip big data is to find high profit areas for taxi drivers. In this paper, we proposed a distributed high profit areas search system which is based on MongoDB and Spark. Our system extract information about high profit areas from raw taxi trips and store them into a MongoDB document store. If a user sends a query for requesting high profit areas, our system returns a set of high profit areas by a distributed skyline processing algorithm with Spark. Experimental results with real-work data demonstrates the feasibility and the effectiveness of our approach.