Opening The Black Box: Explanation And Transparency In Machine Learning