Network :
useful for modelling relationships between entities.
by modelling data as a network we gain an insight into
insights :
important entities : what nodes are important. ie broadcasters or influencers. pathfinding : we can optimize transportation between cities. clustering : levereage nw structure to find communitiies (clustering) within a nw.
nw structure (graph) : nodes edges
Note : node and edges can have metadata associated with them.
networkX (python library)
to manipulate, analyse and model graph data.
import networkx as nx
g = nx.graph()
g.add_nodes_from([1,2,3])
g.nodes()
g.add_edge(1,2) # add egde bewteen 1 and 2.
g.edges() # shows list of edges(tuple of nodes forming edge)
[(1,2)]#Adding a metadata to node
g.node[1]['label'] = 'blue'