Probabilistic Databases of Universal Schema

In data integration we transform information from a source into a target schema. A general problem in this task is loss of fidelity and coverage: the source expresses more knowledge than can fit into the target schema, or knowledge that is hard to fit into any schema at all. This problem is taken to an extreme in information extraction (IE) where the source is natural language. To address this issue, one can either automatically learn a latent schema emergent in text (a brittle and ill-defined task), or manually extend schemas. We propose instead to store data in a probabilistic database of universal schema. This schema is simply the union of all source schemas, and the probabilistic database learns how to predict the cells of each source relation in this union. For example, the database could store Freebase relations and relations that correspond to natural language surface patterns. The database would learn to predict what freebase relations hold true based on what surface patterns appear, and vice versa. We describe an analogy between such databases and collaborative filtering models, and use it to implement our paradigm with probabilistic PCA, a scalable and effective collaborative filtering method