Skip to main content
. Author manuscript; available in PMC: 2010 Jun 11.
Published in final edited form as: Data Min Knowl Discov. 2009 Oct 21;20(3):416–438. doi: 10.1007/s10618-009-0153-2

Table 7.

F-scores for the features that yielded the best prediction results for cancerous vs fractured. As well as the connectivity measures and the Laplacian features the shortest path related measures such average eccentricity and number of central points are important.

Red Clustering Coefficient 1.07
Red Normalized Laplacian Energy 0.94
Red Percentage of Isolated Points 0.81
Red Normalize Laplacian Trace 0.77
Red Average Eccentricity 90 0.76
Red Average Degree 0.71
Blue Giant Connected Ratio 0.67
Blue Clustering Coefficient 0.66
Blue Percentage of Isolated Points 0.61
Blue Percentage of Central Points 0.61