ml-graph-network-analyser

👉 this project contains the source code for the extraction and analysis of several graph (complex networks) features from publicly available datasets with NetworkX

👉 read the final research paper here

---

analyzed features

• assortativity
• clique number
• clustering
• density
• diameter
• edge connectivity
• node connectivity
• number of cliques
• number of edges
• number of nodes
• radius
• clustering and transitivity
• betweenness centrality
• closeness centrality
• communicability centrality
• coreness
• degree centrality
• eccentricity
• number of triangles
• pagerank
• square clustering
• transitivity

---

analyzed networks

• social networks: online social networks, edges represent interactions between people
• ground truth: ground-truth network communities in social and information networks
• communication: email communication networks with edges representing communication
• citation: nodes represent papers, edges represent citations
• collaboration: nodes represent scientists, edges represent collaborations (co-authoring a paper)
• web graphs: nodes represent webpages and edges are hyperlinks
• products: nodes represent products and edges link commonly co-purchased products
• p2p: nodes represent computers and edges represent communication
• roads: nodes represent intersections and edges roads connecting the intersections
• autonomous systems: graphs of the internet
• signed networks: networks with positive and negative edges (friend/foe, trust/distrust)
• location-based networks: Social networks with geographic check-ins
• wikipedia: yalk, editing and voting data from Wikipedia
• bio atlas: food-webs selected from the ecosystem network analysis resources
• bio-cellular: substrate in the cellular network of the corresponding organism
• bio-metabolic: metabolic network of the corresponding organisms
• bio-carbon: carbon exchanges in the cypress wetlands of south florida during the wet and dry season
• bio yeast: protein-protein interaction network in budding yeast

---

additional considerations

normalization and graph sampling

• performed using snowball sampling (choosing the sample order, i.e., number of nodes)
• optimized for the number of edges and multiple samplings

next steps in the data pipeline

• 1. cleanse the data, with ml-netclean
• 2. classify the networks with several machine learning techniques, with mlnet