The computational complexity of edge colouring restricted graphs

dc.contributor.authorCai, Leizhenen_US
dc.date.accessioned2024-08-13T17:19:40Z
dc.date.available2024-08-13T17:19:40Z
dc.date.copyright1988en_US
dc.date.issued1988
dc.degree.departmentDepartment of Computer Science
dc.degree.levelMaster of Science M.Sc.en
dc.description.abstractEdge colouring a graph is a classic algorithmi,; graph problem which has many applications. The problem of determining the chnmatic index is, in general, NP­-complete. This thesis investigates the computational complexity of edge colouring restricted graphs. Previous work has established that the chromatic index problem restricted to k­-regular graphs for any fixed k-3 remains NP-complete, whereas the problem restricted to bipartite graphs and partial k-trees for fixed k is in P. We show that the chromatic index problem restricted to comparability graphs, perfect graphs, line graphs, claw-free graphs, triangle-free graphs and some others remains NP-complete. We present linear time optimal edge colouring algorithms for complete graphs, the liPe graphs of trees and the ln 1e graphs of unicyclic graphs. We also present a linear time approximation algorithm, which uses at most four colours, for cubic graphs, and an O (1.682 IV I) algorithm for determining the chromatic index of any cubic graph.
dc.format.extent89 pages
dc.identifier.urihttps://hdl.handle.net/1828/17396
dc.rightsAvailable to the World Wide Weben_US
dc.titleThe computational complexity of edge colouring restricted graphsen_US
dc.typeThesisen_US

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
CAI_LEIZHEN_MSc_1988_443984.pdf
Size:
1.53 MB
Format:
Adobe Portable Document Format