Which dissimilarity is to be used when extracting typologies in sequence analysis? A comparative study

Sébastien Massoni (1) , Madalina Olteanu (2) , Nathalie N. Villa-Vialaneix (3)

1 CES - Centre d'économie de la Sorbonne
2 UP1 - Université Paris 1 Panthéon-Sorbonne
3 MIAT INRA - Unité de Mathématiques et Informatique Appliquées de Toulouse

Originally developed in bioinformatics, sequence analysis is being increasingly used in social sciences for the study of life-course processes.The methodology generally employed consists in computing dissimilarities between the trajectories and, if typologies are sought, in clustering the trajectories according to their similarities or dissemblances. The choice of an appropriate dissimilarity measure is a major issue when dealing with sequence analysis for life sequences. Several dissimilarities are available in the literature, but neither of them succeeds to become indisputable. In this paper, instead of deciding upon one dissimilarity measure, we propose to use an optimal convex combination of different dissimilarities. The optimality is automatically determined by the clustering procedure and is defined with respect to the within-class variance.

Mots clés: sequence analysis; comparative study; clustering procedure; dissimilarity measure

Mots clés: variance

Domaines: Mathématiques [math]; Informatique [cs]

Sébastien Massoni, Madalina Olteanu, Nathalie N. Villa-Vialaneix. Which dissimilarity is to be used when extracting typologies in sequence analysis? A comparative study. International Workshop on Artificial Neural Networks, Jun 2013, Puerto de la Cruz, Tenerife, Spain. ⟨hal-02806089⟩