Association analysis of location tracking data for various telematics services

Abstract

This paper proposes an approach that extracts the association information from the location data obtained from the real fields but ignored so far. We provide and apply the approach to the real-life location tracking data collected from the Taxi Telematics system developed in Jeju, Korea. The analysis aims at obtaining taxies' meaningful moving patterns for the efficient operations of them. The proposed approach provides the flow chart which would not only take a glance around the overall analysis process but also help save temporal and economic costs required to employ the same or similar data mining analysis to similar services such as public transportations, distribution industries, and so on. Especially, we perform an association analysis on both of refined data and interesting factors extracted from the elementary analysis. The paper proposes the refined association rule mining process as follow: 1) obtaining the integrated dataset through the data cleaning process, 2) extracting the interesting factors from the integrated dataset using frequency and clustering method, 3) performing the association analysis, 4) extracting the meaningful and value-added information such as moving pattern, or 5) returning the feedback to adjust inappropriate factors. The result of the analysis shows that the association analysis makes it possible to detect the hidden moving patterns of vehicles that will greatly improve the quality of Telematics services considering the business requirements. © 2010 Springer-Verlag Berlin Heidelberg.