no code implementations • 29 Nov 2018 • Gaurav Sheni, Benjamin Schreck, Roy Wedge, James Max Kanter, Kalyan Veeramachaneni
In a head-to-head trial, reports generated utilizing full data science automation interface reports were funded 57. 5% of the time, while the ones that used baseline automation were only funded 42. 5% of the time.
1 code implementation • 20 Oct 2017 • Roy Wedge, James Max Kanter, Santiago Moral Rubio, Sergio Iglesias Perez, Kalyan Veeramachaneni
In this paper, we present an automated feature engineering based approach to dramatically reduce false positives in fraud prediction.