PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Représentation et comparaison de séquences par visualisation.
Christine Largeron and Cédric Dreissia
Revue des Nouvelles Technologies de l'Information Number Numéro spécial Visualisation en Extraction des Connaissances, 2005.

Abstract

This paper presents a visual tool for sequences analysis. Sequences are represented by prediction suffix trees (PST). PST can be used to efficiently describe variable order chain. It performs better than the Markov chain of order L and at a lower cost. For this reason, it is interesting to exploit the tree representation not only from a computational point of view but also from a visual one. The tool, developped in this aim, provides a graphic representation of a PST constructed from sequences. It can also be used to compare two models. In supervised classification context, it brings more information than the model by underlying sub-sequences observed in the new sequence and which are not in the model of its predicted class. By the way of visualisation, this tool upgrades the data mining process and the comprehension of the information encoded in sequences.

PDF - Archive staff only - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Information Retrieval & Textual Information Access
ID Code:1574
Deposited By:Christine Largeron
Deposited On:28 November 2005