Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore a mathematical framework for integrating disparate data types that exist on manifolds through this 47-minute conference talk. Learn about Data Integration Via Analysis of Manifolds (DIVAM), an extension of the Data Integration Via Analysis of Subspaces (DIVAS) approach that addresses the challenge of combining different data types in Big Data analysis. Discover how this method handles data objects lying in manifolds, such as shape data, by defining joint variation through modes of variation with identical scores across data blocks. Understand the mathematical foundations for formulating individual variation within each data block using individual modes, and examine how this intrinsic approach extends beyond traditional subspace methods to accommodate the geometric structure of manifold data. Gain insights into advanced statistical techniques for handling complex, high-dimensional data integration problems in modern data science applications.
