We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Farrelly, Gaba, & Kalala Mutombo - Hands-On Network Science | PyData Global 2023
Learn practical network science with Python using NetworkX and iGraph to analyze complex relationships in social networks, epidemics, markets, and more through hands-on examples.
-
Network science provides computationally feasible approaches for analyzing complex data relationships compared to traditional parametric models
-
Key network analysis tools include NetworkX and iGraph, with Python being recommended for beginners
-
Networks can effectively model various real-world scenarios:
- Social networks and relationships
- Disease spread and epidemics
- Stock market behavior and volatility
- Food pricing and supply chains
- Geographic relationships
-
Network analysis benefits include:
- Easy visualization of relationships
- Interpretable results
- Scalable computations
- Less parameter fitting required vs traditional models
-
The Shape of Data and Structure of Complex Network Theory are recommended starting resources for learning network science fundamentals
-
Network structures feature important elements:
- Hubs (densely connected regions)
- Bridges (connections between different groups)
- Weighted vs unweighted connections
- Thresholded relationships
-
Network metrics used for analysis include:
- Forman-Ricci curvature
- Betweenness centrality
- Page rank centrality
- Degree centrality
-
Time series analysis can be enhanced through network approaches:
- Using windowed correlation analysis
- Detecting change points
- Identifying market volatility periods
-
Network science has applications in epidemic control through:
- Modeling disease spread
- Identifying critical nodes for intervention
- Testing containment strategies
-
Future directions include expanding applications in spatio-temporal analysis and scaling solutions for larger datasets