Hover over a node to see the publication title, and to highlight
the publications related to it.
Click on a node to show publication details. It will open up a
On the publication details panel, you can click on the target
icon to center the graph on that movie. And clicking on the link to
a related publication will move to that publication (in the graph and in the
You can use you usual browser controls to zoom and pan
over the graph. If you zoom in when centred on a movie, the
graph will expand around it.
Above a certain zoom level, publication titles are automatically
displayed for all the nodes.
The author names of the participants of the 1st DSML@DBAUG workshop
were extracted, and google scholar was scraped to retrieve the most relevant
results when their names were searched.
Some of the participants did not have a google scholar account and
there was a limit on the maximum number of possible requests on google-scholar.
Nevertheless, 670+ publications were scraped. Also, no care was taken to filter out wrong hits on
the author names (too much work).
When you hover over a node the author name and the neighbors are highlighted.
And how was this concrete "author network" made?
This is how I did it:
Google scholar web scraping (using a Selenium-based library called scholarly) was performed for the author names
The first 20 results were kept (the results contain extended citation info - such as title, abstract, other authors, publication venue & year etc)
The title and abstract were contatenated and a phrase embedding using a pre-trained transformer neural network were computed.
A similarity between all the embeddings of the publications was then computed. This similarity is the edge weight between two nodes.
The visualization of the graph was adapted from this example