Topic Extraction with AGAPE

Authors:
Julien Velcin;Jean-Gabriel Ganascia
Affiliations:
Université de Paris 6 --- LIP6, 104 avenue du Président Kennedy, 75016 Paris,;Université de Paris 6 --- LIP6, 104 avenue du Président Kennedy, 75016 Paris,
Venue:
ADMA '07 Proceedings of the 3rd international conference on Advanced Data Mining and Applications
Year:
2007

Citing 12
Cited 0

Re-designing distance functions and distance-based applications for high dimensional data

ACM SIGMOD Record
Tabu Search

Tabu Search
Cluster validity methods: part I

ACM SIGMOD Record
Knowledge Acquisition Via Incremental Conceptual Clustering

Machine Learning
Information-theoretic co-clustering

Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining
Empirical and Theoretical Comparisons of Selected Criterion Functions for Document Clustering

Machine Learning
A cluster ensembles framework

Design and application of hybrid intelligent systems
A New Conceptual Clustering Framework

Machine Learning
Non-redundant clustering with conditional ensembles

Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining
A Global Optimization RLT-based Approach for Solving the Hard Clustering Problem

Journal of Global Optimization
Probabilistic topic decomposition of an eighteenth-century American newspaper

Journal of the American Society for Information Science and Technology
Stereotype extraction with default clustering

IJCAI'05 Proceedings of the 19th international joint conference on Artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper uses an optimization approach to address the problem of conceptual clustering. The aim of AGAPE, which is based on the tabu-search meta-heuristic using split, merge and a special "k-means" move, is to extract concepts by optimizing a global quality function. It is deterministic and uses no a prioriknowledge about the number of clusters. Experiments carried out in topic extraction show very promising results on both artificial and real datasets.