Deep societal benefits will spring from advances in data availability and in computational procedures for mining insights and inferences from large data sets. I will describe efforts to harness data for making predictions and guiding decisions, touching on work in transportation, healthcare, online services, and interactive systems. I will start with efforts to learn and field predictive models that forecast flows of traffic in greater city regions. Moving from the ground to the air, I will discuss fusing data from aircraft to make inferences about atmospheric conditions and using these results to enhance air transport. I will then focus on experiences with building and fielding predictive models in clinical medicine. I will show how inferences about outcomes and interventions can provide insights and guide decision making. Moving beyond data captured by hospitals, I will discuss the promise of transforming anonymized behavioral data drawn from web services into large-scale sensor networks for public health, including efforts to identify adverse effects of medications and to understand illness in populations. I will conclude by describing how we can use machine learning to leverage the complementarity of human and machine intellect to solve challenging problems in science and society.