马尔科夫逻辑网简介.pptVIP

Domain University of Washington CSE Dept. 12 first-order predicates:Professor, Student, TaughtBy, AuthorOf, AdvisedBy, etc. 2707 constants divided into 10 types:Person (442), Course (176), Pub. (342), Quarter (20), etc. 4.1 million ground predicates 3380 ground predicates (tuples in database) Systems Compared Hand-built knowledge base (KB) ILP: CLAUDIEN [De Raedt Dehaspe, 1997] Markov logic networks (MLNs) Using KB Using CLAUDIEN Using KB + CLAUDIEN Bayesian network learner [Heckerman et al., 1995] Na?ve Bayes [Domingos Pazzani, 1997] Sample Clauses in KB Students are not professors Each student has only one advisor If a student is an author of a paper,so is her advisor Advanced students only TA courses taught by their advisors At most one author of a given paper is a professor Methodology Data split into five areas:AI, graphics, languages, systems, theory Leave-one-area-out testing Task: Predict AdvisedBy(x, y) All Info: Given all other predicates Partial Info: With Student(x) and Professor(x) missing Evaluation measures: Conditional log-likelihood(KB, CLAUDIEN: Run WalkSat 100x to get probabilities) Area under precision-recall curve Results System All Info Partial Info CLL AUC CLL AUC MLN(KB+CL) -0.058 0.152 -0.045 0.203 MLN(KB) -0.052 0.215 -0.048 0.224 MLN(CL) -2.315 0.035 -2.478 0.032 KB -0.135 0.059 -0.063 0.048 CL -0.434 0.048 -0.836 0.037 NB -1.214 0.054 -1.140 0.044 BN -0.072 0.015 -0.215 0.015 Results: All Info Results: Partial Info Efficiency Learning time: 16 mins Time to infer all AdvisedBy predicates: With complete info: 8 mins With partial info: 15 mins (124K Gibbs passes) Other Applications UW-CSE task: Link prediction Collective classification Link-based clustering Social network models Object identification Etc. Other SRL Approaches areSpecial Cases of MLNs Probabilistic relational models(Friedman et al, IJCAI-99) Stochastic logic programs(Muggleton, SRL-00) Bayesian logic programs(Kersting De Raedt, ILP-01) Relational Markov networ