Network Measures Demo

This demo computes structural metrics for a semantic network generated using the Conceptual Network method from fluency data. These measures are useful for comparing networks across individuals, groups, or experimental conditions.

—

Workflow Summary:

  1. Load and flatten fluency data in the “animals” category.

  2. Generate a semantic network using the Conceptual Network method.

  3. Convert the resulting adjacency matrix to a NetworkX graph.

  4. Compute common network metrics using built-in NetworkX functions.

  5. Save the results to a .pkl file for further use.

—

Metrics Calculated:

  • clustering_coefficient: How often a node’s neighbors are connected.

  • density: Ratio of edges to all possible edges in the graph.

  • number_of_edges: Total connections in the network.

  • number_of_nodes: Vocabulary size represented in the graph.

  • average_node_degree: Average degree of nodes, based on neighbors’ degree.

  • average_shortest_path_length: Average number of steps in shortest paths (largest component only).

  • diameter: Longest shortest path (largest component only).

These measures are calculated using both the full network and the largest connected component, which avoids failures in disconnected graphs.

—

Functions and Tools Used:

  • snafu.conceptualNetwork

  • networkx.Graph

  • nx.average_clustering, nx.density, nx.diameter, etc.

  • pickle.dump() for result storage

Output:

  • cn_metrics_expected.pkl: A serialized Python dictionary of the computed metrics.