Detecting and quantifying clusters in graphs is a highly generalizable problem useful in a myriad of scenarios (e.g. understanding collective behaviour, large-image processing, etc.). Yet there is currently no implementation in GraphX for community detection. That's where I step in.
Biomedical research is a rich field of study encompassing many area of multidisciplinary expertise. This tool aims to be a stepping stone towards empowering academics and medical staff for better decision making. Discovering new collaboration and obtaining a clearer understanding of crucial social links between entities are on the table.
By making this code publicly available, I hope to help others solving similar problems. Contributing this implementation to the GraphX team by merging with the main branch is also envisioned.
Personal and community-wide statistics on growth regarding coauthorship of scientific publication.
How are you positioned inside your community of peers? How could this affect your proficiency?
Very large graphs are difficult to visualize in any interactive manner. By providing a hierarchical exploration of nodes we allow the user to navigate the complexity.
Observing trends at finer granularity of your population allows crucial insights to be gathered at the right resolution.