Toward Normative Expert Systems: Part I The Pathfinder Project

By David E. Heckerman, Eric Horvitz, B. Nathwani. 1992

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Improving the accuracy of medical diagnosis with causal machine learning (2020, methodology): “BN disease models have a long history going back to the INTERNIST-1 18 , Quick Medical Reference (QMR) 19,20 , and PATHFINDER 21,22 systems, with many of the original systems corresponding to noisy-OR networks with only disease and symptom nodes, known as BN2O networks 36 .”
Bayesian Networks and Decision Graphs (2001, methodology): “A version of probability updating in singly connected DAGs through message passing was presented by Kim and Pearl (1983). In (Pearl 1986), conditioning was used to reduce propagation in multiply connected networks to propagation in singly connected networks. Shachter (1986) introduces arc reversal a”
A characterization of Markov equivalence classes for acyclic digraphs (1997, methodology): “15For example, Spiegelhalter et al (1993, Section 3) and Heckerman et al (1992) model dependencies in biomedical data by means of ADGs with n = 20 and n = 108 vertices, respectively.”
Qualitative Verbal Explanations in Bayesian Belief Networks (1996, background): “, [10, 18, 20]), which is the logarithm of the likelihood ratio describing the in uence of the nding on the variable of interest.”
Cyberchondria: Studies of the escalation of medical concerns in Web search (2009, methodology): “There is an opportunity to detect if a searcher is employing Web search to perform diagnosis, much as (s)he might use a medical expert system [Heckerman et al. 1992] if such a system were available, where symptoms are input and a list of a reasonable explanations ranked by their likelihood is…”
Learning Sparse Nonparametric DAGs (2019, background): “Learning DAGs from data is an important and classical problem in machine learning, with a diverse array of applications in causal inference (Spirtes et al., 2000), fairness and accountability (Kusner et al., 2017), medicine (Heckerman et al., 1992), and finance (Sanford and Moosa, 2012).”
Learning Bayesian networks: approaches and issues (2011, methodology): “Another more specific diagnostic system comes from the Pathfinder project (Heckerman et al., 1992), which is used in the diagnosis of lymph-node diseases.”
Capabilities of GPT-4 on Medical Challenge Problems (2023, background): “…by shifting enthusiasm for diﬀerent representations and reasoning methods, including core probabilistic and decision-theoretic methods (e.g., [GB68, HHN92]), rule-based production systems (e.g., [Sho77, BS84]), semantic graphs (e.g., [PSS81]), supervised learning from databases of medical…”