Object-oriented data analysis (OODA) is a framework for analyzing datasets with much richer structure than the usual numbers or vectors. Examples of object data include shapes, functions, images and networks.  At the start of any OODA application we need to answer the following questions: What are the data objects? What is the object representation space? What is the feature space for carrying out practical statistical analysis? What statistical methods will be used?  This talk by FIU Professor Dr. Ian Dryden will focus on shape analysis, which is of interest in many practical applications. Typically the representation space of shapes is a quotient space, where translation, rotation and size have been removed. An important aspect of OODA is to reduce the dimension to a small number key features while respecting the geometry of the manifold in which objects lie.

Two case studies are given with objects at big and little scales:

• Describing face shape, and its relationship with age and gender
• Clustering the shapes of dynamic proteins.

This event is being held both in-person and through a zoom stream.