Partitioning Graphs into Monochromatic Trees Problems
Abstract
Graph partition problems is an important class of graph problems that are able to model real-world tasks such as Concurrent Processor Allocation and Very Large-Scale Integration (VLSI) systems. Even though such problems are NP-hard, there is a lot of effort in developing methods to deal with such problems. In this work, we present results from the point of view of computational complexity for the problem of Partitioning Graphs into Monochromatic Trees as know as PGMT, proving the NP-hardness even for restricted versions of this problem. In addition, we present a lower bound for the execution time of algorithms based on the Exponential Time Hypothesis (ETH) and we also present two algorithms, the first based on dynamic programming when considering that the input graph is a tree and a second algorithm parameterized by the treewidth and the number of colors.
References
Jin, Z., Kano, M., Li, X., and Wei, B. (2006). Partitioning 2-edge-colored complete multipartite graphs into monochromatic cycle, paths and trees. Journal of Combinatorial Optimization, 11(4):445–454.
Jin, Z. and Li, X. (2004). The complexity for partitioning graphs by monochromaic trees, cycles and paths. International Journal of Computer Mathematics, 81(11):1357–1362.
Jin, Z. and Li, X. (2008). Vertex partitions of r-edge-colored graphs. Applied Mathematics-A Journal of Chinese Universities, 23(1):120–126.
