We propose a framework for building graph-ical causal model that is based on the con-cept of causal mechanisms. Causal models are intuitive for human users and, more im-portantly, support the prediction of the effect of manipulation. We describe an imple-mentation of the proposed framework as an interactive model construction module, Ima-GeNIe, in SMILE (Structural Modeling, In-ference, and Learning Engine) and in GeNIe (SMILE's Windows user interface).
展开▼