Localized Multiple Sources Self-Organizing Map

Ludovic Platon (1) , Farida Zehraoui (1) , Fariza Tahi (1)

1 IBISC - Informatique, Biologie Intégrative et Systèmes Complexes

We present in this paper a new approach based on unsupervised self organizing maps called MSSOM. This approach combines multiple heterogeneous data sources and learns the weights of each source at the level of clusters instead of learning the same source weights for the whole space. This allows to improve the performances of our model especially in applications where a local feature selection is important. We evaluate our method using several artificial and real datasets and show competitive results compared to the state-of-art.

Mots clés: Dimensionality reduction; Kernel methods; Neural network; Self-Organizing Map; Unsupervised learning

Domaines: Bio-informatique [q-bio.QM]; Apprentissage [cs.LG]

Ludovic Platon, Farida Zehraoui, Fariza Tahi. Localized Multiple Sources Self-Organizing Map. 25th International Conference on Neural Information Processing (ICONIP 2018), Dec 2018, Siem Reap, Cambodia. pp.648--659, ⟨10.1007/978-3-030-04182-3_57⟩. ⟨hal-01971022⟩