Yuval Moskovitch

In today's data-driven world, insights collected from the data and trends observed in the data significantly contribute to decision making. However, users are often perplexed by certain surprising data trends, especially the disparate ones. For example, upon observing a disparate trend that "men are more likely to have a heart-attack than women", a health-care professional wonders, "is there a certain demographic where the trend is more pronounced or even reversed?", "what factors further exacerbate or alleviate such disparity?". To this end, we introduce ExDis, a system for automatically identifying data regions where an observed Disparity is pronounced (or reversed) and Explaining the associated causes that exacerbate (or alleviate) the disparity. ExDis equips policymakers to recognize the factors that causally contribute to certain disparities and implement targeted corrective measures.