## Probabilistic Frame-Based Systems (1998)

### Cached

### Download Links

- [www.eecs.harvard.edu]
- [www.aaai.org]
- [www.aaai.org]
- [www.aaai.org]
- [www.aaai.org]
- DBLP

### Other Repositories/Bibliography

Venue: | In Proc. AAAI |

Citations: | 197 - 18 self |

### BibTeX

@INPROCEEDINGS{Koller98probabilisticframe-based,

author = {Daphne Koller and Avi Pfeffer},

title = {Probabilistic Frame-Based Systems},

booktitle = {In Proc. AAAI},

year = {1998},

pages = {580--587},

publisher = {AAAI Press}

}

### Years of Citing Articles

### OpenURL

### Abstract

Two of the most important threads of work in knowledge representation today are frame-based representation systems (FRS's) and Bayesian networks (BNs). FRS's provide an excellent representation for the organizational structure of large complex domains, but their applicability is limited because of their inability to deal with uncertainty and noise. BNs provide an intuitive and coherent probabilistic representation of our uncertainty, but are very limited in their ability to handle complex structured domains. In this paper, we provide a language that cleanly integrates these approaches, preserving the advantages of both. Our approach allows us to provide natural and compact definitions of probability models for a class, in a way that is local to the class frame. These models can be instantiated for any set of interconnected instances, resulting in a coherent probability distribution over the instance properties. Our language also allows us to represent important types of uncertainty tha...

### Citations

7436 |
Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference
- Pearl
- 1988
(Show Context)
Citation Context ... decade by the development of Copyright 1998, American Association for Artificial Intelligence (www.aaai.org). All rights reserved. probabilistic reasoning systems, and particularly Bayesian networks =-=[10]-=-. A Bayesian network (BN) is a representation of a full joint distribution over a set of random variables; it can be used to answer queries about any of its variables given any evidence. A BN allows a... |

498 |
Building Large Knowledgebased Systems
- Lenat, Guha
- 1989
(Show Context)
Citation Context ...o use that information in answering probabilistic queries. 1 Introduction Frame representation systems (FRS's) are currently the primary technology used for large scale knowledge representation in AI =-=[8, 3, 7]-=-. Their modular organization according to cognitively meaningful entities and their ability to capture patterns common to many individuals provide a convenient language for representing complex struct... |

438 | The Ontolingua Server: A Tool for Collaborative Ontology Construction
- Farquhar, Fikes, et al.
- 1996
(Show Context)
Citation Context ...o use that information in answering probabilistic queries. 1 Introduction Frame representation systems (FRS's) are currently the primary technology used for large scale knowledge representation in AI =-=[8, 3, 7]-=-. Their modular organization according to cognitively meaningful entities and their ability to capture patterns common to many individuals provide a convenient language for representing complex struct... |

302 | Context-specific independence in Bayesian networks
- Boutilier, Friedman, et al.
- 1996
(Show Context)
Citation Context ...alue of I:B. The value of I:B:oe is equal to the value of K[ae]:oe. In other words, I:ref (B) selects the value of I:B:oe from a set of possibilities. Therefore, the node I:B:oe is a multiplexer node =-=[1]-=-; it has as parents the node I:ref (B) and all nodes K[ae]:B:oe, and it uses the value of I:ref (B) to select, as its value, the value of one of its appropriate parents. ProcessIndirect(I:B:) AddParen... |

302 | Probabilistic Horn abduction and bayesian networks
- Poole
- 1993
(Show Context)
Citation Context ...ng standard for FRS interoperability. Our work is a signficant improvement over previous approaches to combining first-order logic and Bayesian networks. Most of the attempts in this direction (e.g., =-=[12, 11, 9]-=-) use probabilistic Horn clauses as the basic representation. The choice of Horn clauses as an underlying language already dictates some of the properties of the representation, e.g., its inability to... |

195 |
Propagating Uncertainty Bayesian Networks by Probabilistic Logic Sampling
- Henrion
- 1988
(Show Context)
Citation Context ...f, during the sampling process, a value is generated for a slot which is inconsistent with the observed value, we simply discard the entire partial value generated up to that point. It is easy to see =-=[4]-=- that the relative probability with which a partial value is generated in this data generating process is exactly the same as its probability conditioned on the observed slot values. As we discussed, ... |

190 | Object-oriented Bayesian networks
- Koller, Pfeffer
- 1997
(Show Context)
Citation Context ...bility of BNs is largely limited to situations that can be encoded, in advance, using a fixed set of attributes. Thus, they are inadequate for large-scale complex KR tasks. Buildingon our recent work =-=[6, 5]-=-, we propose a representation language that integrates frame-representation systems and Bayesian networks, thereby providing the first bridge between these two very different threads of work in KR. Th... |

187 |
The evolving technology of classification-based knowledge representation systems
- MacGregor
- 1991
(Show Context)
Citation Context ...o use that information in answering probabilistic queries. 1 Introduction Frame representation systems (FRS's) are currently the primary technology used for large scale knowledge representation in AI =-=[8, 3, 7]-=-. Their modular organization according to cognitively meaningful entities and their ability to capture patterns common to many individuals provide a convenient language for representing complex struct... |

108 | P-CLASSIC: A tractable probabilistic description logic
- Koller, Levy, et al.
- 1997
(Show Context)
Citation Context ...bility of BNs is largely limited to situations that can be encoded, in advance, using a fixed set of attributes. Thus, they are inadequate for large-scale complex KR tasks. Buildingon our recent work =-=[6, 5]-=-, we propose a representation language that integrates frame-representation systems and Bayesian networks, thereby providing the first bridge between these two very different threads of work in KR. Th... |

95 | Answering queries from contextsensitive probabilistic knowledge bases
- Ngo, Haddawy
- 1996
(Show Context)
Citation Context ...ng standard for FRS interoperability. Our work is a signficant improvement over previous approaches to combining first-order logic and Bayesian networks. Most of the attempts in this direction (e.g., =-=[12, 11, 9]-=-) use probabilistic Horn clauses as the basic representation. The choice of Horn clauses as an underlying language already dictates some of the properties of the representation, e.g., its inability to... |

64 |
From knowledge bases to decision models
- Wellman, Breese, et al.
- 1992
(Show Context)
Citation Context ...ng standard for FRS interoperability. Our work is a signficant improvement over previous approaches to combining first-order logic and Bayesian networks. Most of the attempts in this direction (e.g., =-=[12, 11, 9]-=-) use probabilistic Horn clauses as the basic representation. The choice of Horn clauses as an underlying language already dictates some of the properties of the representation, e.g., its inability to... |

3 |
Open knowledge base connectivity 2.0.2. Available from http://www.ai.sri.com/˜okbc
- Chaudhri, Farquhar, et al.
- 1998
(Show Context)
Citation Context ...ystem, and thereby to annotate existing KBs with probabilistic information. In particular, we have implemented a system based on our approach, capable of interacting with most existing FRS's via OKBC =-=[2]-=-, an emerging standard for FRS interoperability. Our work is a signficant improvement over previous approaches to combining first-order logic and Bayesian networks. Most of the attempts in this direct... |

3 |
ProbabilisticReasoning in Intelligent Systems
- Pearl
- 1988
(Show Context)
Citation Context ... decade by the development of Copyright 1998, American Association for Artificial Intelligence (www.aaai.org). All rights reserved. probabilistic reasoning systems, and particularly Bayesian networks =-=[10]-=-. A Bayesian network (BN) is a representation of a full joint distribution over a set of random variables; it can be used to answer queries about any of its variables given any evidence. A BN allows a... |