The study of complex networks pervades all of the science. We can assign complex networks into four general classes (although there is some overlap between them): technological networks (e.g., Internet, the telephone network, power grids, transportation network), information networks (e.g., the world wide web, citation networks), biological networks (e.g., biochemical network, neutral networks, ecological networks), and social networks.
Characterizing complex network's structure is a key to understand any unifying principles underlying their topology. Several previous works have shown that many topological properties can vary for different types of system. However these works generally focus only a few characteristics at time. In this project, we present the first part of a method to characterize complex networks by performing an extensive analysis of the global and local topological features of networks. In a second part, these features are used into input vectors for a SVN classifier, establishing an efficient way of learning the classification of complex networks.
To use this software you can extract the data and calculate the features with [this repository] (https://github.com/mariwahl/NetAna-Complex-Network-Analysis)
And cleanse the data with [this repository] (https://github.com/mariwahl/NetClean-Complex-Networks-Data-Cleanser)
The feature vectors were extracting using MNet in this repository
We have vectors for different normalization (Snowball and Metropolis Hastings Random Walk samplings) for different sizes. We also have vectors for the entire graphs for some of the features (that were possible to be calculated).
These vectors are parsed and cleansed using [this repository] (https://github.com/mariwahl/NetClean-Complex-Networks-Data-Cleanser)
We perform classification of the network features using many classifiers:
- SVM (supervised)
- Logistic Regression (supervised)
- Adaboost (supervised)
- EM (unsupervised)
The comparisons of the the many classifiers and the plots are available under each classifier's folder.
Drop me a line or submit a patch!
We obtain excellent results for our classifiers:
- SVM RESULTS, (train/test PERCENTAGE: 0.8)
linear xmin, atrain: 0.815, atest: 0.809
SVC xmin, atrain: 0.734, atest: 0.739
linear gauss, atrain: 0.821, atest: 0.816
SVC gauss, atrain: 0.916, atest: 0.914
linear none, atrain: 0.6, atest: 0.611
SVC none, atrain: 0.998, atest: 0.993 ---> ALMOST 100% ACCURARY!!!!!
- ADABOOST RESULTS, (train/test PERCENTAGE: 0.8)
tech, xmin , atrain: 0.906114845197 , atest: 0.905554614733
info, xmin , atrain: 0.961694628209 , atest: 0.95816045724
social, xmin , atrain: 0.932485842816 , atest: 0.924731160034
bio, xmin , atrain: 0.977130457793 , atest: 0.979229466554
tech, gauss , atrain: 0.904228631913 , atest: 0.905944115157
info, gauss , atrain: 0.959289229955 , atest: 0.955270956816
social, gauss , atrain: 0.930746229161 , atest: 0.925279424217
bio, gauss , atrain: 0.977036782218 , atest: 0.979362828112
- LOGISTIC REGRESSION, (train/test PERCENTAGE: 0.8)
xmin, atrain: 0.821, atest: 0.814
gauss, atrain: 0.827, atest: 0.823 ---> 83% ACCURACY
none, atrain: 0.745, atest: 0.748
Type Siz Ord Ass Tra Deg Cor NTr NCl Cnu Clu Eco Ecc Dia Bet Den Rad Scl Com Pag Cen
xmin 0.145 0.545 1.0 0.34 0.845 0.585 0.655 0.0 0.32 1.0 0.0 0.505 0.465 0.715 0.97 0.54 0.45 0.0 1.0 1.0
gauss 0.155 0.52 1.0 0.365 0.785 0.605 0.585 0.0 0.415 1.0 0.0 0.51 0.48 0.7 0.96 0.57 0.41 0.0 1.0 1.0
none 0.155 0.565 1.0 0.3 0.845 0.55 0.6 0.0 0.48 1.0 0.0 0.5 0.495 0.69 0.97 0.475 0.47 0.0 1.0 1.0