Probabilistic Datalog---a logic for powerful retrieval methods

ing with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Publications Dept, ACM Inc. , fax +1 (212) 869-0481, or permissions@acm. org 2 \ ---Since Datalog P allows for recursive rules, it provides more powerful inference than any other (implemented) probabilistic IR model. ---Finally, since Datalog P is a generalization of (deterministic) Datalog, it can be used as a standard query language for both IR and database systems, and thus also for integration of these two types of systems on the logical level. 2. INFORMAL DESCRIPTION OF Datalog P Probabilistic Datalog is an extension of stratified Datalog (see e. g. Ullman 88, Ceri et al. 90). On the syntactical level, the only difference is that with ground facts, also a probabilistic weight may be given, e. g. 0. 7 indterm (d1, ir). 0. 8 indterm (d1, db). Informally speaking, the probabilistic weight gives. . .