Radio Frequency Identification (RFID) technology gains momentum and every day we witness more and more application domains. These systems however generate a huge volume of data and magement of such vast volume of data is a challenge. It is even bigger challenge to extract patterns and relations, which represent a knowledge that is implicitly stored in such spatio-temporal data. This knowledge cannot be obtained using a simple queries. While a significant work has been devoted toward general data mining a limited work was directed toward mining of spatio-temporal RFID data. In this work, we present method which efficiently cluster vast volume of RFID data by applying K-means method only to fraction of whole data covering all locations. In empirical study we show that our method is efficient and at the same time has a property of high accuracy.

Presented at Conferences

  • IASTED International Conference on Artificial Intelligence and Applications (AIA 2009) (2009)

    Innsbruck, Austria