Use NetworkRepository data

Advanced: assumes basic Python, tabular data, and graph queries. For a guided introduction, start with Your first graph.

NetworkRepository is an external collection of network datasets. Start with Your first graph before adapting one. GraphForge v0.6.0 does not ship a NetworkRepository downloader, GraphML loader, or catalog convenience API.

Inspect the source before analysis

Choose an individual dataset page and follow its original-study citation. Check the actual download format instead of assuming every file is GraphML. Record node and edge meanings, direction, weights, collection period, source version, and applicable use terms.

A co-purchase link, a reported friendship, and a shared appearance in a chapter are different observations. A community-detection result does not by itself explain motivations, social identity, or causation. Read the study and any qualitative material alongside the network.

Construct the graph

Parse the downloaded format with an appropriate external tool, then use Graph construction to create nodes and relationships through GraphForge’s public API. The data preparation guide covers source identifiers, missing records, and checks before analysis. There are no verified catalog load-time or automatic-cache guarantees here.

For a runnable small example, use the visualization guide. It uses Zachary’s Karate Club from Mark Newman’s network data collection, with an explicit source and checksum. That example does not qualify every NetworkRepository dataset or replace reading the original research.

Retain source files and preprocessing decisions so another reader can reproduce your graph. Use save and reopen for persistence.