Clustering dense graphs: A web site graph paradigm

Moussiades, Lefteris; Vakali, Athena

doi:10.1016/j.ipm.2009.11.003

Cited by 6 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, in step 4, a graph-clustering algorithm may be used to reveal the clusters of plagiarism. A lot of graph clustering algorithms have been proposed [22,23,24,25,26,27,28,29,30,31,32,18,20]. However, most of them require a prespecified number of clusters as input.…”

Section: Description Of Pdamentioning

confidence: 99%

Discovering Clusters of Plagiarism in Students’ Source Codes

Moussiades¹

2016

JESTR

Self Cite

View full text Add to dashboard Cite

Plagiarism in students' source codes constitutes an important drawback for the educational process. In addition, plagiarism detection in source codes is time consuming and tiresome task. Therefore, many approaches for plagiarism detection have been proposed. Most of the aforementioned approaches receive as input a set of source files and calculate a similarity between each pair of the input set. However, the tutor often needs to detect the clusters of plagiarism, i.e. clusters of students' assignments such as all assignments in a cluster derive from a common original. In this paper, we propose a novel plagiarism detection algorithm that receives as input a set of source codes and calculates the clusters of plagiarism. Experimental results show the efficiency of our approach and encourage us to further research.

show abstract

Section: Description Of Pdamentioning

confidence: 99%

Discovering Clusters of Plagiarism in Students’ Source Codes

Moussiades¹

2016

JESTR

Self Cite

View full text Add to dashboard Cite

show abstract

“…The value of parameter in Algorithms 2 and 3 takes constant 12, which is an approximate parameter value in many nearest neighbors related algorithm in practice. The SNN similarity threshold in the Algorithm 2 is selected from [1,12]. In Algorithm 3, the neighborhood radius is selected from [1,10] and the least point parameter MinPts is selected from [1,10].…”

Section: Parameters Settingmentioning

confidence: 99%

“…The SNN similarity threshold in the Algorithm 2 is selected from [1,12]. In Algorithm 3, the neighborhood radius is selected from [1,10] and the least point parameter MinPts is selected from [1,10].…”

Section: Parameters Settingmentioning

confidence: 99%

“…When the researchers try to use an objective measure to a precise definition of class, they often find that the optimal clustering problem cannot be calculated at all. At present, the existing full-dimensional clustering methods can be roughly divided into these kinds: partition method [1][2][3][4], hierarchical method [5][6][7][8], density-based method [9,10], grid-based method [11] and on-line clustering [12,13]. On clustering high-dimensional data, the popular methods have been roughly divided into these kinds: subspace clustering, pattern-based clustering, and correlation clustering [11,14,15].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Smooth Splicing: A Robust SNN-Based Method for Clustering High-Dimensional Data

Tan

Wang

2013

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Sharing nearest neighbor (SNN) is a novel metric measure of similarity, and it can conquer two hardships: the low similarities between samples and the different densities of classes. At present, there are two popular SNN similarity based clustering methods: JP clustering and SNN density based clustering. Their clustering results highly rely on the weighting value of the single edge, and thus they are very vulnerable. Motivated by the idea of smooth splicing in computing geometry, the authors design a novel SNN similarity based clustering algorithm within the structure of graph theory. Since it inherits complementary intensity-smoothness principle, its generalizing ability surpasses those of the previously mentioned two methods. The experiments on text datasets show its effectiveness.

show abstract

“…Graph-clustering algorithm plays an important role in many fields: monitoring on computer network executive purpose, visualization knowledge based on support of understanding of complex data structure, measurement data cluster, detection source code plagiarism, network data cluster and online community identification [3,5,6,7] . Clustering based on graph theory is one of the clustering method developed rapidly recently, which is based on graph theory and computer graphics [1,2,4,8] .…”

Section: Introductionmentioning

confidence: 99%

A Graph Clustering Algorithm Based on the Intra Vertex Adjacent Ratio

Guo¹,

Wang²

2016

Proceedings of the 2016 International Conference on Applied Mathematics, Simulation and Modelling

View full text Add to dashboard Cite

This paper proposes a new clustering coefficient based on the number of connecting vertices within a cluster (called inter vertex adjacent ratio, IVAR) and proposes a new clustering algorithm based on IVAR. Finally, the analysis of algorithm applicability shows that the algorithm is more applicative and effective for regular graphs than the vertex-clustering algorithm based on intra connection ratio [L. Moussiades and A. Vakali, Clustering dense graph: A web site graph paradigm. Information Processing and Management, vol. 46, pp.247-267, 2010].

show abstract

Clustering dense graphs: A web site graph paradigm

Cited by 6 publications

References 30 publications

Discovering Clusters of Plagiarism in Students’ Source Codes

Discovering Clusters of Plagiarism in Students’ Source Codes

Smooth Splicing: A Robust SNN-Based Method for Clustering High-Dimensional Data

A Graph Clustering Algorithm Based on the Intra Vertex Adjacent Ratio

Contact Info

Product

Resources

About