查看文章

The field of computer science has greatly benefitted from the advances in combina- torial algorithms in the last few decades. This is because the advent of high speed computers has made it possible to generate lists of combinatorial objects in a practi- cal amount of time. In areas such as genome science and data mining the problems are often vaguely defined, and researchers have to look for meaningful information in huge datasets. In this thesis, efficient generation algorithms for subgraphs such as bicliques and paths in cliques are developed. Cliques and bicliques are used to model various real-world problems encountered in bio-informatics, data mining and networks. We consider two variations of bicliques: pseudo-bicliques and c-isolated bicliques. Pseudo- bicliques relax the rigid connectivity requirement of bicliques to cater for missing and noisy data. On the other hand, the c-isolated bicliques enforce a …