Extraction of Commuter Behavior on Commuter Line Using Rule-Based Machine Learning

Abstract

The application of Automatic Fare Collection (AFC) on Commuter Line trains can provide new knowledge in navigating between Commuter Line train lines and real commuter travel data. The AFC system allows management to obtain large amounts of detailed data regarding the routes of each commuter daily. One of the challenges faced in using big data at AFC is the extraction of data on the behavior of transporting passengers. Commuter Line passenger behavior is a very important factor for operators to make the right decision. This study uses the association rules method to extract AFC data to produce good information and understand Jabodetabek commuter behavior. The results showed that the association rules method could extract AFC data and produce strong association rules on commuter behavior.