Dynamics in Graph Analysis Adding Time as a Structure for Visual and Statistical
summary
Claims that embedding time structurally reduces traversal cost and clarifies dynamics are demonstrated with concrete tools and counts: NetworkX and graph_tool (C++ with Python API) plus baline and minky process ~700,000 documents, an initial 2.6 million node graph sampled to 139,000 then filtered to ~8,500 nodes and 112,000 edges. The talk builds both the add time as a node model and the multiple subgraphs snapshots model, shows week granularity, compares centrality measures over time, and produces animated layouts that highlight what actually changes.
AI-written summary · about
chapters (26)
- 0:00Graph analytics tools intro
- 1:47Graphs for data science
- 3:56Graph algorithms benefits
- 6:05Human intuition in graphs
- 7:29Four graph algorithm classes
- 9:15Traversal core operation
- 10:16Property graph structure
- 13:47Time in relational databases
- 15:33Time in property graphs
- 17:43Time-bounded graph traversal
- 18:24Email network time analysis
- 20:29Subgraphing with time filters
- 22:14Dynamic graph methods
- 23:39Time as graph element
- 25:03NLP pipeline baline minky
- 27:08Key phrase filtering
- 27:49Power law in graph data
- 28:09Ego graph extraction
- 28:53Time centrality in graphs
- 31:00Centrality over time
- 32:24News cycles visualization
- 33:48Dynamic graph visual analytics
- 34:29Graph layout techniques
- 35:56Visual analytics mantra
- 36:39Graphs for analytics
- 40:10Q&A
related talks
38mDynamics in Graph Analysis: Adding Time as a Structure for Visual and Statisti
31mGraph Analytics - From the Whiteboard to Your Toolbox
01h 26mBuilding an Interactive Network Graph to Understand Communities
48mBig Graph: Big Data applied to giant, dynamic graphs
43mVisualising networks
27mConnecting the Dots: Graph Analytics for Economic Sustainability
02h 47mNetwork Analysis Made Simple
28mNetworkX is Fast Now: Graph Analytics Unleashed