CRSOM is a ncRNAs classifier using heterogeneous sources. CRSOM combines self-organizing maps (SOM), which is an unsupervised learning algorithm, with a multilayer peceptrons (MLP) that is a supervised learning algorithm. CRSOM learns the source weights locally at the level of a cluster and uses rejection options in order to improve the reliability of the predictions and to identify potential new classes.


CRSOM folder


For any questions, comments or suggestions about CRSOM, please feel free to contact: fariza.tahi@ibisc.univ-evry.fr