Social Network Analysis

Integer programming formulations of discrete hub location problems

[서평]「Algorithms on Strings, Trees, and Sequences」

: Significant progress has been made with solution of location problems and in preprocessing and decomposition for discrete optimization. There has also been research on the application of techniques from combinational optimization to nonlinear problems.

http://www.dtic.mil/dtic/tr/fulltext/u2/a273552.pdf

Discrete Applied Mathematics

Clustering by fast search and find of density peaks via heat diffusion

Finding cohesive subgroups is an important issue in studying social networks. Many models exist for defining cohesive subgraphs in social networks, such as clique, $$k$$ -clique, and $$k$$ -clan. The concept of $$k$$ -club is one of them. A $$k$$ -club of a graph is a maximal subset of the vertex set which induces a subgraph of diameter $$k$$ . It is a relaxation of a clique, which induces a subgraph of diameter $$1$$ . We conducted algorithmic studies on finding a $$k$$ -club of size as large as possible. In this paper, we show that one can find a $$k$$ -club of maximum size in $$O^{*}(1.62^n)$$ time where $$n$$ is the number of vertices of the input graph. We implemented a combinatorial branch-and-bound algorithm that finds a $$k$$ -club of maximum size and a new heuristic algorithm called IDROP given in this paper. To speed up the programs, we introduce a dynamic data structure called $$k$$ -DN which, under deletion of vertices from a graph, maintains for a given vertex $$v$$ the set of vertices at distances at most $$k$$ . From the experimental results that we obtained, we concluded that a $$k$$ -club of maximum size can be easily found in sparse graphs and dense graphs. Our heuristic algorithm finds, within reasonable time, $$k$$ -clubs of maximum size in most of experimental instances. The gap between the size of a $$k$$ -club of maximum size and a $$k$$ -club found by IDROP is a constant for the number of vertices that we are able to test.

http://par.cse.nsysu.edu.tw/~algo/paper/paper11/A1_1.pdf

Finding large $$k$$-clubs in undirected graphs

Fixed-parameter algorithms for vertex cover P3

MapReduce is a crucial framework in the cloud computing architecture, and is implemented by Apache Hadoop and other cloud computing platforms. The resources required for executing jobs in a large data center vary according to the job types. In general, there are two types of jobs, CPU-bound and I/O-bound, which require different resources but run simultaneously in the same cluster. The default job scheduling policy of Hadoop is first-come-first-served and therefore, may cause unbalanced resource utilization. Considering various job workloads, numerous job allocation schedulers were proposed in the literature. However, those schedulers encountered the data locality problem or unreasonable job execution performance. This study proposes a job scheduler based on a dynamic grouping integrated neighboring search strategy, which can balance the resource utilization and improve the performance and data locality in heterogeneous computing environments.

Ling-Ju Hung

Papers

Finding large $$k$$-clubs in undirected graphs

Fixed-parameter algorithms for vertex cover P3

Heterogeneous Job Allocation Scheduler for Hadoop MapReduce Using Dynamic Grouping Integrated Neighboring Search

An improved algorithm for the maximum agreement subtree problem

Exact algorithms for problems related to the densest k-set problem