HU facial recognition software predicts criminality

A group of Harrisburg University of Science and Technology students and professors have developed an automated computer classifier capable of predicting with 80% accuracy and no racial bias whether someone is likely to be a criminal based solely on a picture of their face.
Jonathan W. Korn, a PhD student in Harrisburg University of Science and Technology’s Data Science program and a NYPD veteran; Prof. Nathaniel J.S. Ashby, and Prof. Roozbeh Sadeghian’s research titled “A Deep Neural Network Model to Predict Criminality Using Image Processing” will appear in the forthcoming Springer Nature – Research Book Series: Transactions on Computational Science & Computational Intelligence

Read more

Consumer Stockpiling During COVID-19 Crisis Can Look Panicky, But It Has Its Rational Side

Consumers are clearing store shelves. Some observers call it “panic buying.” But a Johns Hopkins University expert on consumer behavior, while acknowledging that panic is an element of the phenomenon, says stockpiling can be seen as a rational approach to shopping during a pandemic.

Read more