Demonstration of Inferring Causality from RelationalDatabases with CaRL


Understanding cause-and-effect is key for informed decision-making. �The gold standard in causal inference is performing controlled �experiments, �which �may �not �always �be �feasible due to ethical, legal, or cost constraints. �As an alternative, inferring causality from observational data has been extensively �used �in �statistics �and �social �sciences. � However, �the existing methods critically rely on a restrictive assumption that the population of study consists of homogeneous units that can be represented as a single flat table. �In contrast, in many real-world settings, the study domain consists of heterogeneous units that are best represented using relational databases. � We �propose �and �demonstrateCaRL: �an �end-to-end system for drawing causal inference from relational data. �In addition, we built a visual interface to wrap aroundCaRL. In our demonstration, we will use this GUI to show a �live �investigation �of �causal �inference �from �real �academic and medical relational databases.

To appear in VLDB 2020