TY - GEN

T1 - Better decomposition heuristics for the maximum-weight connected graph problem using betweenness centrality

AU - Yamamoto, Takanori

AU - Bannai, Hideo

AU - Nagasaki, Masao

AU - Miyano, Satoru

PY - 2009/11/16

Y1 - 2009/11/16

N2 - We present new decomposition heuristics for finding the optimal solution for the maximum-weight connected graph problem, which is known to be NP-hard. Previous optimal algorithms for solving the problem decompose the input graph into subgraphs using heuristics based on node degree. We propose new heuristics based on betweenness centrality measures, and show through computational experiments that our new heuristics tend to reduce the number of subgraphs in the decomposition, and therefore could lead to the reduction in computational time for finding the optimal solution. The method is further applied to analysis of biological pathway data.

AB - We present new decomposition heuristics for finding the optimal solution for the maximum-weight connected graph problem, which is known to be NP-hard. Previous optimal algorithms for solving the problem decompose the input graph into subgraphs using heuristics based on node degree. We propose new heuristics based on betweenness centrality measures, and show through computational experiments that our new heuristics tend to reduce the number of subgraphs in the decomposition, and therefore could lead to the reduction in computational time for finding the optimal solution. The method is further applied to analysis of biological pathway data.

UR - http://www.scopus.com/inward/record.url?scp=71049152085&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=71049152085&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-04747-3_40

DO - 10.1007/978-3-642-04747-3_40

M3 - Conference contribution

AN - SCOPUS:71049152085

SN - 3642047467

SN - 9783642047466

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 465

EP - 472

BT - Discovery Science - 12th International Conference, DS 2009, Proceedings

T2 - 12th International Conference on Discovery Science, DS 2009

Y2 - 3 October 2009 through 5 October 2009

ER -