e[discover] objectively and statistically measures the value of data assets as defined by their usefulness and relevance. The approach is based on decision theory and exploits quantitative and probabilistic techniques to enable a unique conversational/question driven approach, allowing scientists to explore scientific value across diverse data sets as never previously possible.
e[discover] won the prestigious Best of Show Award at Bio-IT World Conference & Expo 2016. Find out more by watching this award video. This features an interview with Eagle Genomics’ CEO, Abel Ureta-Vidal and VP Raminderpal Singh.
How it works
e[discover] is a first-generation tool capable of objectively assessing data value (as opposed to “quality” or monetary value) of scientific datasets using Decision Theory techniques. A hierarchical model is used to match datasets (value components) to the question posed by the data scientist. Once modelling is complete, e[discover] allows the user to apply value to the catalogue, discover relationships between components contributing to the value, and to further refine the model.
Unique question-driven approach » allows scientists to explore datasets using validated scientific queries.
Objective guide to data curation and enhancement »
Data selection by prioritization (scoring) » not simply down-selection (filtering).
Hierarchical search & discovery visualization »
Enterprise integration with proprietary or public data sources »
Translation of tacit knowledge to a business experimental process » subject to PDSA learning cycle
Conversational environment (human-machine & human-human) » for refining data value model