Integration Strategy Is a Key Step in Network-Based Analysis and Dramatically Affects Network Topological Properties and Inferring Outcomes

Table 4

The topological properties of the 11 new empirical and machine learning networks.

	Empirical				Machine learning
	Union	Intersection	2-Vote	3-Vote	Naive Bayes	Bayesian Networks	Logistic Regression	SVM	Random Tree	Random Forest	J48
Proteins	14936	507	9548	5558	14840	14869	14890	14895	14486	14860	14570
Interactions	145534	497	40766	12891	134095	140956	140746	142226	114598	139082	120541
Diameter	15	6	16	15	15	15	16	16	16	15	16
Degree	19.49	1.96	8.54	4.64	18.07	18.96	18.90	19.10	15.82	18.72	16.55
Density	0.00130	0.00387	0.00089	0.00083	0.00122	0.00128	0.00127	0.00128	0.00109	0.00126	0.00114
ASP	2.9216	1.9164	4.4487	4.7394	2.9372	2.9245	2.9256	2.9217	3.0447	2.9280	3.0103
CC	0.0206	0.1262	0.0471	0.0340	0.0161	0.0204	0.0197	0.0204	0.0156	0.0194	0.0170

Note: Proteins, Interactions, Diameter, Degree, and Density indicate the number of proteins, the number of interactions, the network diameter, the average degree and the network density, respectively. ASP and CC are the average path length and clustering coefficient, respectively. Bold type indicates the minimum value for average path length and the maximum value for the other topological properties of the empirical and machine learning methods.