Toward Normative Expert Systems: Part II Probability-Based Representations for Efficient Knowledge Acquisition and Inference. Heckerman, D. E & Nathwani, B. N Methods of Information in Medicine, 1992. abstract bibtex We address practical issues concerning the construction and use of decision-theoretic or normative expert systems for diagnosis. In particular, we examine Pathfinder, a normative expert system that assists surgical pathologists with the diagnosis of lymph-node diseases, and discuss the representation of dependencies among pieces of evidence within this system. We describe the belief network, a graphical representation of proba-bilistic dependencies. We see how Pathfinder uses a belief network to construct differen-tial diagnoses efficiently, even when there are dependencies among pieces of evidence. In addition, we introduce an extension of the belief-network representation called a similarity network, a tool for constructing large and complex belief networks. The rep-resentation allows a user to construct independent belief networks for subsets of a given domain. A valid belief network for the entire domain can then be constructed from the individual belief networks. We also introduce the partition, a graphical representation that facilitates the assessment of probabilities associated with a belief network. Finally, we show that the similarity-network and partition representations made practical the construction of Pathfinder.
@article{heckerman_toward_1992,
title = {Toward {Normative} {Expert} {Systems}: {Part} {II} {Probability}-{Based} {Representations} for {Efficient} {Knowledge} {Acquisition} and {Inference}},
abstract = {We address practical issues concerning the construction and use of decision-theoretic or normative expert systems for diagnosis. In particular, we examine Pathfinder, a normative expert system that assists surgical pathologists with the diagnosis of lymph-node diseases, and discuss the representation of dependencies among pieces of evidence within this system. We describe the belief network, a graphical representation of proba-bilistic dependencies. We see how Pathfinder uses a belief network to construct differen-tial diagnoses efficiently, even when there are dependencies among pieces of evidence. In addition, we introduce an extension of the belief-network representation called a similarity network, a tool for constructing large and complex belief networks. The rep-resentation allows a user to construct independent belief networks for subsets of a given domain. A valid belief network for the entire domain can then be constructed from the individual belief networks. We also introduce the partition, a graphical representation that facilitates the assessment of probabilities associated with a belief network. Finally, we show that the similarity-network and partition representations made practical the construction of Pathfinder.},
journal = {Methods of Information in Medicine},
author = {Heckerman, David E and Nathwani, Bharat N},
year = {1992},
}
Downloads: 0
{"_id":"ZC32h7GH9oFQtf6uZ","bibbaseid":"heckerman-nathwani-towardnormativeexpertsystemspartiiprobabilitybasedrepresentationsforefficientknowledgeacquisitionandinference-1992","downloads":0,"creationDate":"2017-04-19T15:50:25.606Z","title":"Toward Normative Expert Systems: Part II Probability-Based Representations for Efficient Knowledge Acquisition and Inference","author_short":["Heckerman, D. E","Nathwani, B. N"],"year":1992,"bibtype":"article","biburl":"https://bibbase.org/zotero/davidrpoa","bibdata":{"bibtype":"article","type":"article","title":"Toward Normative Expert Systems: Part II Probability-Based Representations for Efficient Knowledge Acquisition and Inference","abstract":"We address practical issues concerning the construction and use of decision-theoretic or normative expert systems for diagnosis. In particular, we examine Pathfinder, a normative expert system that assists surgical pathologists with the diagnosis of lymph-node diseases, and discuss the representation of dependencies among pieces of evidence within this system. We describe the belief network, a graphical representation of proba-bilistic dependencies. We see how Pathfinder uses a belief network to construct differen-tial diagnoses efficiently, even when there are dependencies among pieces of evidence. In addition, we introduce an extension of the belief-network representation called a similarity network, a tool for constructing large and complex belief networks. The rep-resentation allows a user to construct independent belief networks for subsets of a given domain. A valid belief network for the entire domain can then be constructed from the individual belief networks. We also introduce the partition, a graphical representation that facilitates the assessment of probabilities associated with a belief network. Finally, we show that the similarity-network and partition representations made practical the construction of Pathfinder.","journal":"Methods of Information in Medicine","author":[{"propositions":[],"lastnames":["Heckerman"],"firstnames":["David","E"],"suffixes":[]},{"propositions":[],"lastnames":["Nathwani"],"firstnames":["Bharat","N"],"suffixes":[]}],"year":"1992","bibtex":"@article{heckerman_toward_1992,\n\ttitle = {Toward {Normative} {Expert} {Systems}: {Part} {II} {Probability}-{Based} {Representations} for {Efficient} {Knowledge} {Acquisition} and {Inference}},\n\tabstract = {We address practical issues concerning the construction and use of decision-theoretic or normative expert systems for diagnosis. In particular, we examine Pathfinder, a normative expert system that assists surgical pathologists with the diagnosis of lymph-node diseases, and discuss the representation of dependencies among pieces of evidence within this system. We describe the belief network, a graphical representation of proba-bilistic dependencies. We see how Pathfinder uses a belief network to construct differen-tial diagnoses efficiently, even when there are dependencies among pieces of evidence. In addition, we introduce an extension of the belief-network representation called a similarity network, a tool for constructing large and complex belief networks. The rep-resentation allows a user to construct independent belief networks for subsets of a given domain. A valid belief network for the entire domain can then be constructed from the individual belief networks. We also introduce the partition, a graphical representation that facilitates the assessment of probabilities associated with a belief network. Finally, we show that the similarity-network and partition representations made practical the construction of Pathfinder.},\n\tjournal = {Methods of Information in Medicine},\n\tauthor = {Heckerman, David E and Nathwani, Bharat N},\n\tyear = {1992},\n}\n\n","author_short":["Heckerman, D. E","Nathwani, B. N"],"key":"heckerman_toward_1992","id":"heckerman_toward_1992","bibbaseid":"heckerman-nathwani-towardnormativeexpertsystemspartiiprobabilitybasedrepresentationsforefficientknowledgeacquisitionandinference-1992","role":"author","urls":{},"metadata":{"authorlinks":{}},"html":""},"search_terms":["toward","normative","expert","systems","part","probability","based","representations","efficient","knowledge","acquisition","inference","heckerman","nathwani"],"keywords":[],"authorIDs":[],"dataSources":["M66iSiMeC2pAhHPfr","fvS7tt2THEKMbWDXR"]}