search

WDscholia / scholia

Wikidata-based scholarly profiles
https://scholia.toolforge.org
Other
223 stars 80 forks source link

New visualisations #1108

Open TS404 opened 4 years ago

TS404 commented 4 years ago

I've been experimenting with some visualisations not directly based on the default wikidata query service but that draw from wikidata queries in [R] (Code examples, R notebook as html, related issue). It may be worth considering moving to D3 visualisations eventually to allow for more customisable visualisations and display layout? Interactive chord diagram example: image image

fnielsen commented 4 years ago

We could do this more. I made the "work-topic matrix visualization" in the author aspect. This is the only do far.

TS404 commented 4 years ago

I've found that in both this and network diagrams benefit from colouring by cluster (either clustering by label propagation [1] or DBSCAN [2]). It also allows for logical orering of nodes around the perimieter in a chord diagram.

  1. Raghavan, U.N. and Albert, R. and Kumara, S.: Near linear time algorithm to detect community structures in large-scale networks. Phys Rev E 76, 036106. (2007)
  2. Hahsler M, Piekenbrock M, Doran D (2019). dbscan: Fast Density-Based Clustering with R. Journal of Statistical Software, 91(1), 1-30. doi: 10.18637/jss.v091.i01
fnielsen commented 4 years ago

We already serve D3 https://github.com/fnielsen/scholia/tree/master/scholia/app/static so it is "just" the Javascript glue between WDQS and D3 that is needed.

TS404 commented 4 years ago

Interesting. I've been having a go at a co-topic network using networkD3 in [R]. It works well locally, but I'm still working on getting the website to function properly:

carlinmack commented 3 years ago

Glad to hear that we are using D3 currently, will try to close #458 with it :)

Adafede commented 2 years ago

https://github.com/XavierGimenez/vega-transform-sparql maybe of interest for implementing such visualizations?