Label Propagation for Semi-Supervised Learning in Self-Organizing Maps

o Semi-supervised learning aims at discov- ering spatial structures in high-dimensional input spaces when insufcient background information about clusters is available. A particulary interesting approach is based on propagation of class labels through proximity graphs. The Emergent Self-Organizing Map (ESOM) itself can be seen as such a proximity graph that is suitable for label propagation. It turns out that Zhu's popular label prop- agation method can be regarded as a modication of the SOM's well known batch learning technique. In this pa- per, an approach for semi-supervised learning is presented. It is based on label propagation in trained Emergent Self- Organizing Maps. Furthermore, a simple yet powerful method for crucial parameter estimation is presented. The resulting clustering algorithm is tested on the fundamental clustering problem suite (FCPS).