Evaluation of text clustering methods and their dataspace embeddings: an exploration
Résumé
Fair evaluation of text clustering methods needs to clarify the relations between 1)pre-processing, resulting in raw term occurrence vectors, 2)data transformation, and 3)method in the strict sense. We have tried to empirically compare a dozen well-known methods and variants in a protocol crossing three contrasted open-access corpora in a few tens transformed dataspaces. We compared the resulting clusterings to their supposed "ground-truth" classes by means of four usual indices. The results show both a confirmation of well-established implicit combinations, and good performances of unexpected combinations, mostly in spectral or kernel dataspaces. The rich material resulting from these some 450 runs includes a wealth of intriguing facts, which needs further research on the specificities of text corpora in relation to methods and dataspaces.
Mots clés
method comparison
Cluster Analysis
Clustering Methods
Text Clustering
Text Mining
K-Means
Graph Clustering
NMF
Non-negative Matrix Factorization
Latent Dirichlet Allocation
LDA
hierarchical clustering
linkage method
spectral clustering
graph partition
kernel clustering
Normalized Matrix Factorization
Statistical Evaluation
Benchmark
Okapi
BM25
Correspondence Analysis
chi-square
tf-idf
Reuters' ModApté Split
ACM collection
20 Newsgroups collection
polynomial kernel
Gram matrix
Evaluation method
Domaines
Informatique [cs]
Fichier principal
Cadot-Lelu_Cluster-Challenge7.pdf (1.14 Mo)
Télécharger le fichier
Links_to_codes.pdf (56.82 Ko)
Télécharger le fichier
PresentationIFCS19LeLu-Cadot_ccc.pdf (995.56 Ko)
Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)