Meeting 38: Topological Data Analysis with Mapper

Please join us for meeting #38 where we'll discuss two papers describing and applying Mapper for topological data analysis. Topological data analysis is an approach that extends established machine-learning with ideas from topology, a field until recently regarded as the purest of math. We found an accessible entry point into the area, where it meshes with earlier ideas on clustering, and produces interesting and unexpected results.

G. Singh, F. Memoli, G. Carlsson. Topological Methods for the Analysis of High Dimensional Data Sets and 3D Object Recognition (https://research.math.osu.edu/tgda/mapperPBG.pdf).

P. Y. Lum, G. Singh, A. Lehman, T. Ishkanov, M. Vejdemo-Johansson, M. Alagappan, J. Carlsson & G. Carlsson. Extracting insights from the shape of complex data using topology (http://www.nature.com/articles/srep01236) (we'll skip the supplement which is not public).

No previous knowledge of topology will be required.

If time permits, we will also discuss Mapper's open source implementation by Daniel Muellner, http://danifold.net/mapper (a powerful research-grade code with impressive visual output.)

Please read the papers before the meetup, but if you don't get a chance, feel free to come join the discussion anyway.

Past papers are listed in the message board here (https://www.meetup.com/Silicon-Valley-Data-Science-Journal-Club/messages/boards/thread/40111312).

Post new paper suggestions here (https://www.meetup.com/Silicon-Valley-Data-Science-Journal-Club/messages/boards/thread/40111062).