How does one search for a needle in a multi-dimensional haystack not knowing what a needle is and not being sure there is one in the haystack? Solving this sort of problem might seem to be impossible, yet this is exactly what Dynamic Quantum Clustering (DQC) manages to do. Several key features of DQC are: it is unbiased in that it makes no assumptions about the type, structure or number of clusters that might exist; it is data agnostic, in that it uses no domain-specific knowledge; it doesn’t find clusters when presented with random data; it works when other clustering methods fail to work. These advantages mean that DQC works well for data coming from fields as different from one another as biology, physics, medicine, finance and even national security.
Dr. Weinstein's talk will cover examples drawn from many successful applications of DQC. In each of these, conventional clustering methods failed to produce useful results. The examples are real data from a wide variety of disciplines including x-ray absorption spectroscopy, earthquake science, particle physics, condensed matter physics and biology. These vary in size from thousands to millions entries. They convincingly demonstrate DQC’s power and flexibility.