Randomization techniques for statistical significance testing on graphs
Sami Hanhijärvi, Gemma Garriga and Kai Puolamäki
In: 6th International Workshop on Mining and Learning with Graphs (MLG '08), 4-5 July 2008, Helsinki, Finland.

Abstract

Studying the patterns and properties of graph data is important in many application areas. A crucial question remains still largely ignored: how significant are the data mining results found on the graph data? Currently, the results are mostly justified by the optimal or near optimal value of the defined objective function. We study randomization techniques for testing the statistical significance of graph analysis results.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

[Edit]