Ref HAL: hal-00628587_v1
DOI: 10.1007/978-3-540-88067-7_16
Résumé: The goal of any clustering algorithm producing flat partitions of data, is to find both the optimal clustering solution and the optimal number of clusters. One natural way to reach this goal without the need for parameters, is to involve a validity index in a clustering process, which can lead to an objective selection of the optimal number of clusters. In this chapter, we provide two main contributions. Firstly, since validity indices have been mostly studied in a two or three-dimensionnal datasets, we have chosen to evaluate them in a real-world applications, document and word clustering. Secondly, we propose a new context-aware method that aims at enhancing the validity indices usage as stopping criteria in agglomerative algorithms. Experimental results show that the method is a step-forward in using, with more reliability, validity indices as stopping criteria.

Commentaires: sci09shz

Résumé: The aim of this book is to gather the most recent works that address issues related to the concept of mining complex data. The whole knowledge discovery process being involved, our goal is to provide researchers dealing with each step of this process by key entries. Actually, managing complex data within the KDD process implies to work on every step, starting from the pre-processing (e.g. structuring and organizing) to the visualization and interpretation (e.g. sorting or filtering) of the results, via the data mining methods themselves (e.g. classification, clustering, frequent patterns extraction, etc.). The papers presented here are selected from the workshop papers held yearly since 2006. The book is composed of four parts and a total of sixteen chapters. Part I gives a general view of complex data mining by illustrating some situations and the related complexity. It contains five chapters. Chapter 1 illustrates the problem of analyzing the scientific literature. The chapter gives some background to the various techniques in this area, explains the necessary pre-processing steps involved, and presents two case studies, one from image mining and one from table identification.

Commentaires: sci09ztrh

DOI: 10.1007/978-3-540-68123-6_53
Résumé: The goal of any clustering algorithm producing flat partitions of data, is to find both the optimal clustering solution and the optimal number of clusters. One natural way to reach this goal without the need for parameters, is to involve a validity index in a clustering process, which can lead to an objective selection of the optimal number of clusters. In this paper, we provide an evaluation of the major relative indices involving them in an agglomerative clustering algorithm for documents. The evaluation seeks the indices ability to identify both the optimal solution and the optimal number of clusters. Then, we propose a new context-aware method that aims at enhancing the validity indices usage as stopping criteria in agglomerative algorithms. Experimental results show that the method is a step-forward in using, with more reliability, validity indices as stopping criteria.

Commentaires: ismis08shz2

Commentaires: ismis08shz

DOI: 10.1007/978-3-540-68125-0_103
Résumé: Ontology learning from text is considered as an appealing and a challenging approach to address the shortcomings of the hand-crafted ontologies. In this paper, we present OLEA, a new framework for ontology learning from text. The proposal is a hybrid approach combining the pattern-based and the distributional approaches. It addresses key issues in the area of ontology learning: low recall of the pattern-based approach, low precision of the distributional approach, and finally ontology evolution. Preliminary experiments performed at each stage of the learning process show the pros and cons of the proposal.

Commentaires: pakdd08shz