New and exciting:

I’m featured in a Washington Post article and video about testing autonomous cars

These are articles I’ve written and projects I’ve created that cover electronic discovery (software seeking the smoking gun), data science, and machine learning.

Plus One: an app I wrote (in R) that predicts the next word based on what you enter (up to four words), along the lines of what your phone offers when you’re texting. It was trained on tweets and news articles from about five years ago, and it won’t swear.

Transparent Concept Search, a feature I designed that lets users search using “turkey,” and get back documents that are about Grandma and pumpkin pie, or about issues in the Middle East.

Can lawyers be replaced by software that can analyze legal documents?: I worked in the electronic discovery (eDiscovery) field for a long time, which is essentially software for lawyers handling extremely large civil cases. One of the big discussions was whether or not using machine learning classification should be allowed to decide which documents would be turned over to the other side as part of discovery. (I’m on the “yes” side.)

Is there a cure for MS?: This is a very exciting new use for an existing treatment for leukemia. After having this treatment, 78% of the MS patients in a major study had (so far) permanent remission.