Topographic Connectionist Unsupervised Learning for RFID Behavior Data Mining

Abstract

Radio Frequency IDentification (RFID) is an advanced tracking technology that can be used to study the spatial organization of animal societies. The aim of this work is to build a new RFID-based autonomous system to follow individuals spatio-temporal activity, which is not currently available, and to develop new tools for automatic data mining. We study here how to transform these data to obtain knowledge about the division of labor and intra-colonial cooperation and conflict in an ant colony by developing a new unsupervised learning data mining method (DS2L-SOM : Density-based Simultaneous Two-Level Self Organizing Map) to find homogeneous clusters (i.e., sets of individual witch share a distinctive behavior). This method is very fast and efficient and it also allows a very useful visualization of results.