AntClust: Ant Clustering and Web Usage Mining

In this paper, we propose a new ant-based clustering algorithm called AntClust. It is inspired from the chemical recognition system of ants. In this system, the continuous interactions between the nestmates generate a "Gestalt" colonial odor. Similarly, our clustering algorithm associates an object of the data set to the odor of an ant and then simulates meetings between ants. At the end, artificial ants that share a similar odor are grouped in the same nest, which provides the expected partition. We compare AntClust to the K-Means method and to the AntClass algorithm. We present new results on artificial and real data sets. We show that AntClust performs well and can extract meaningful knowledge from real Web sessions.

[1]  Marco Dorigo,et al.  Distributed Optimization by Ant Colonies , 1992 .

[2]  B. Hölldobler,et al.  The kin recognition system of carpenter ants (Camponotus spp.) , 1986, Behavioral Ecology and Sociobiology.

[3]  Nicolas Monmarché,et al.  A new clustering algorithm based on the chemical recognition system of ants , 2002 .

[4]  Nicolas Monmarché,et al.  On how Pachycondyla apicalis ants suggest a new search algorithm , 2000, Future Gener. Comput. Syst..

[5]  Baldo Faieta,et al.  Diversity and adaptation in populations of clustering ants , 1994 .

[6]  Jeffrey Heer,et al.  Mining the Structure of User Activity using Cluster Stability , 2002 .

[7]  Pascale Kuntz,et al.  Emergent colonization and graph partitioning , 1994 .

[8]  Patrick Brézillon,et al.  Lecture Notes in Artificial Intelligence , 1999 .

[9]  Jeffrey Heer,et al.  Identification of Web User Traffic Composition using Multi-Modal Clustering and Information Scent , 2000 .

[10]  Vladimir Estivill-Castro,et al.  Categorizing Visitors Dynamically by Fast and Robust Clustering of Access Logs , 2001, Web Intelligence.

[11]  D. Cliff From animals to animats 3 : proceedings of the Third International Conference on Simulation of Adaptive Behavior , 1994 .

[12]  Lawrence W. Lan,et al.  Genetic clustering algorithms , 2001, Eur. J. Oper. Res..

[13]  Nicolas Monmarché,et al.  On Improving Clustering in Numerical Databases with Artificial Ants , 1999, ECAL.

[14]  B. Hölldobler,et al.  The kin recognition system of carpenter ants (Camponotus spp.) , 1987, Behavioral Ecology and Sociobiology.

[15]  Umeshwar Dayal,et al.  From User Access Patterns to Dynamic Hypertext Linking , 1996, Comput. Networks.

[16]  R. Matthews,et al.  Ants. , 1898, Science.