Title: NEW CROSSING MINIMIZATION TECHNIQUE FOR CANCER DATASETS BICLUSTERING

Year of Publication: 2013
Page Numbers: 248-253
Authors: Ahmed Sharaf Eldein, Samar Kassim, Tamer Mohamed
Conference Name: The Third International Conference on Digital Information Processing and Communications (ICDIPC2013)
- United Arab Emirates

Abstract:


Cancer is a fatal disease that causes large number of deaths all over the world; so many studies have been made on genes to study the behavior of this disease. Biclustering is one of the data mining techniques that aiming to find the behavior of subset of genes under subset of circumstances. In this paper we propose a new biclustering algorithm that uses the Barycenter value of each node and its position in the graph beside the weight and position of the neighbors of each node the graph to give a new weight for each node and reordering the nodes depending on this weight. The ordering of bipartite graph grouping the related nodes together which enhances the results of biclustering algorithm.