Given the random nature of sports events, it's understandably hard to accurately project player performance in any given contest, especially in the NFL, with such a limited set of games to draw from. This presentation outlines a method to improve upon existing fantasy player projections published by professional sites (IE NFL.com) by incorporating sentiment based features into the projection model, and discusses results achieved with this approach.
Bringing together the leading figures in sports analytics, business, and technology.
.svg){kind=small}
