PNNL's Signature Discovery team has developed a formal process for signature discovery that transforms signature development to become

• More efficient by reducing trial-and-error and decreasing the time to discovery a signature
• More economical by delivering methods and tools that allow users to reuse, rather than reinvent, the resources needed to construct, detect, and validate signatures
• More rigorous by providing robust and well-defined processes for signature discovery

Challenges

Modern signatures often involve features from many different data types. Signatures are most useful when they are easily interpreted by decision-makers.

Key questions

• How do we best select features from multiple measurement sources to construct our signature?
• How do we detect signatures in "real world" environments?
• How do we assess the quality of a signature and compare different signatures?
• How do we recognize change and adapt signatures to dynamic phenomena?
• Is this process generalizable across domains?

 [[{"type":"media","view_mode":"media_large","fid":"215","attributes":{"alt":"","class":"media-image","height":"480","typeof":"foaf:Image","width":"235"}}]]