Efficient Data Mining Method to Localise Errors in RFID Data

Abstract

Since the emergence of Radio Frequency Identification technology (RFID), the community has been promised a cost effective and efficient means to identify and track large number of items with relative ease. Unfortunately, due to the unreliable nature of the passive architecture, the RFID revolution has been reduced to a fraction of intended audience due to the anomalies. These anomalies are duplicate, positive and negative readings. While duplicate readings and wrong data (false positive) can be easily identified and rectified, that is not the case for false negative or missed readings. To identify missed readings data mining methods can be used. However, due to its vast volume and complex spatio-temporal structure of RFID data, traditional data mining methods are not necessarily directly applicable. In this paper we propose method to identify possible missed RFID readings by applying association rules data mining method. In empirical study we show that our algorithm is accurate and efficient and also we show that it scales well with increased number of rows therefore it is applicable on vast volume on spatio-temporal RFID data.