DELVE is a prototype for performing literature-based searches with the aid of interactive visualizations and a framework for quickly implementing such visualizations as modular Web-applications.
Clouds indicate the frequency of common words, phrases, or MeSH terms. Word trees show the context of key-phrases. A list of documents is updated according to what the user has focused upon. Future planned applications will include phrase nets, publication history analysis, and topic model analysis.
For proof of concept, we implemented DELVE using PubMed citations made available by the NLM. Currently, many search engines, including PubMed, return a linear list of search results. A DELVE search returns instead an interactive set of panes through which different visualizations enable insightful data exploration; researchers delve into their query so that relevant results more quickly enter into their vision and irrelevant results can be systematically removed and dismissed.
DELVE is a collaboration between individuals at the Center for Clinical and Translational Science at the University of Kentucky and the School of Biomedical Informatics at the University of Texas Health Science Center at Houston. Please read more about us.