Modeling text with generalizable Gaussian mixtures

Author

Hansen, Lars ; Sigurdsson, Sigurdur ; Kolenda, Thomas ; Nielsen, Finn Årup ; Kjems, Ulrik ; Larsen, Jan

Author_Institution

Dept. of Math. Modelling, Tech. Univ. Denmark, Lyngby, Denmark

Volume

6

fYear

2000

fDate

2000

Firstpage

3494

Abstract

We apply and discuss generalizable Gaussian mixture (GGM) models for text mining. The model automatically adapts model complexity for a given text representation. We show that the generalizability of these models depends on the dimensionality of the representation and the sample size. We discuss the relation between supervised and unsupervised learning in the test data. Finally, we implement a novelty detector based on the density model

Keywords

Gaussian processes; computational complexity; information retrieval; pattern recognition; unsupervised learning; Web browser; Web visualization; density model based detector; generalized Gaussian mixtures; information retrieval; model complexity; pattern recognition; representation dimension; sample size; supervised learning; test data; test modeling; text mining; text representation; unsupervised learning; Cost function; Detectors; Feature extraction; Histograms; Information retrieval; Large scale integration; Mathematical model; Pattern recognition; Statistics; Supervised learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2000. ICASSP '00. Proceedings. 2000 IEEE International Conference on

Conference_Location

Istanbul

ISSN

1520-6149

Print_ISBN

0-7803-6293-4

Type

conf

DOI

10.1109/ICASSP.2000.860154

Filename

860154