Over the past decade, there have been lots of advances in life sciences. One of these came from the combination of insights in molecular science and genetics and the huge progress in computer science. Cartographer visualises the trends that lie within the UCB interest sphere. From a research perspective, the goal to reveal and monitor new and upcoming trends was of significant importance to us.
XAOP started from abstracts and metadata from 5.5 million Pubmed articles and applied text mining, processing and machine learning. Pubmed, the leading library containing abstracts and links to medical, nursing, health care and preclinical sciences journal articles, was used as the main data source. To create a quick and powerful overview, we combined different data sources, such as Pubmed, with NIH data of grants and several patent databases. This mix enabled us to reveal trend evolutions in predefined life science interest fields.
The main component is a self-exploratory data visualisation. To visually represent hierarchy, XAOP chose to use a horizontal tree diagram. The starting point or the root node is always an interest sphere. And from there on out, the user just needs to click and scroll through the connections to discover related terms and cross domain trends. Each time the user clicks on a topic, he/she goes deeper into the diagram. The relevance and evolution over time has not been forgotten. With one simple click, you get an overview of the selected topic/term from 2012-2015.
If you’d like to know a bit more about this project, make sure to check out our blogpost Predicting, monitoring and visualising trends in life sciences written by Elke Van Assche, Frauke Demol and Nicolas Teirlinckx.