Approximate Graphlets Extraction
Input format: tab separated weighted edgelist (nodes represented with integer ids).
src trg w
node_id0 node_id1 weight
Input format: tab separated nodelist (nodes represented with integer ids).
node_id0 nlabel
The algorithm can be used as standalone program as well as integrated in python scripts.
python AGraphlets.py network_filename percentile backbone_threshold min_graphlet_size max_graphlet_size -a node_attr_file
where:
- network_filename: edgelist filename
- percentile: component filtering percentile in [0, 100]
- backbone_threshold: edge statistical significance threshold in [0,1]
- min_graphlet_size: minimum size for graphlet patterns
- max_graphlet_size: maximum size for graphlet patterns
- node_attr_file: node labels filename (optional)
import AGraphlets as a
ag = a.AGraphlet(network_file, node_attr=node_attr, approx_percentile=percentile, backbone_threshold=backbone_threshold)
ag.execute(min_pattern_size=min_graphlet_size, max_pattern_size=max_graphlet_size)