1 The Geometric Structure of Data

You might wonder why you need data science methods rooted in topology and geometry when traditional data science methods are already so popular and powerful. The answer to this has two parts. First, data today comes in a variety of formats far more exotic than the usual spreadsheet, such as a social network or a text document. While such exotic data used to be referred to as unstructured, we now recognize that it often is structured, but with a more sophisticated geometric structure than a series of spreadsheets in a relational database. Topological and geometric data science allow us to work directly in ...

Get The Shape of Data now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.