Digital humanities Filtering
Filtering Filtering exercise Chinese/English Buddhists study keyword co-occurrence 1987~ Compare modularity analysis before and after filtering Giant component Clustering coefficient Edge weights Frequency
modularity analysis before filtering Run
Select
Results: Modularity: 0.651 Modularity with resolution: 0.651 Number of Communities: 312
Notice: under Node Overview Clustering coefficient analysis Run Notice: under Node Overview Not under Dynamic
Choose undirected
Results: Average Clustering Coefficient: 0.827
Filter Queries 1.Frequency 2.Edge weights 3.Clustering coefficient 4. Giant component Queries
Giant Component Topology Giant Component “ Drag filter here” under Queries
Clustering coefficient Attributes Range Clustering coefficient “ Drag subfilter here” under Giant Component Open
Clustering coefficient click Type 0.09
“ Drag subfilter here” under Clustering coefficient Edge Weight Edges edge whight “ Drag subfilter here” under Clustering coefficient Type in 0.6
Frequency Attributes Range total “ Drag subfilter here” under edge whight Type in 3
After filtering Filter
modularity analysis after filtering Results: Modularity: 0.875 Modularity with resolution: 0.875 Number of Communities: 24
Also Presidential speech Compare modularity analysis before and after filtering Try filter insignificant terms and come up with a better modularity analysis Use Marvel social network and aNobii book co-ownership , Movie viewing data to play with filtering, centrality, and modularity analysis
An example of high frequency and low clustering coefficient term “Buddhists”