Inspiration
Extracting actionable information from complex data may not be straightforward. Standard tools applied directly to data which is highly dimensional, or when noise is present can result in inaccurate and unreliable conclusions. For more accurate outcomes one strategy is to transform data into signatures that are less sensitive to the noise and irrelevant signals. There is a growing body of evidence from road network reconstructions, neuroscience, vehicle tracking, object recognition, protein structure understanding, suggesting effectiveness of topology based signatures. The aim of this course is to present how to extract topology based signatures and use them in practice.
See some of the final projects of this course at Grades and projects page.
See the following videos for inspiration about Topological Data Analysis: