GeNIe & SMILE - Network Repository

Characteristics:

Description:
Decision models from Clemen's book 'Making Hard Decisions'.

Reference:
Robert T. Clemen, Making Hard Decisions: An Introduction to Decision Analysis, Second Edition

Characteristics:

Nodes count:		37
Mean indegree		1.24
Max. indegree:		4
Avg. outcomes count:		2.84
Max. outcomes count:		4

Description:
A network by medical experts for monitoring patients in intensive care.

Reference:
I. Beinlich and G. Suermondt and R. Chavez and G. Cooper, The ALARM monitoring system: A case study with two probabilistic inference techniques for belief networks, Proceedings of the 2nd European Conference on AI and Medicine, Springer-Verlag, Berlin, 1989.

ZIP Andes

Characteristics:

Nodes count:		223
Mean indegree		1.5157
Max. indegree:		6
Avg. outcomes count:		2
Max. outcomes count:		2

Description:
ANDES is an intelligent tutoring system for classical Newtonian physics that was being developed by a team of researchers at the Learning Research and Development Centerat the University of Pittsburgh and researchers at the United States Naval Academy. The student model in ANDES uses a Bayesian network to do long-term knowledge assessment, plan recognition, and prediction of students' actions during problem solving.

Reference:
@InProceedings{Conati97,
author = "Conati, C. and Gertner, A. S. and VanLehn, K. and Druzdzel, M. J.",
title = "On-line Student Modeling for Coached Problem Solving Using {B}ayesian Networks",
booktitle="Proceedings of the Sixth International Conference on User Modeling (UM--96)",
publisher="Springer Verlag",
address = "Vienna, New York",
year = 1997,
pages = "231--242"
}

ZIP Barley

Characteristics:

Nodes count:		48
Mean indegree		1.75
Max. indegree:		4
Avg. outcomes count:		8.77
Max. outcomes count:		67

Description:
A network by medical experts for monitoring patients in intensive care.

Reference:
Preliminary model for barley developed under the project : "Production of beer from Danish malting barley grown without the use of pesticides" http://www.agrsci.dk/plb/uikr/projects/maltingbarley_uk.shtml by Kristian Kristensen , Ilse A. Rasmussen and others.

ZIP DT Tutor IDx

Characteristics:

Chance nodes count:		149
Decision nodes count:		2
Deterministic nodes count:		12
Utility nodes count:		35
Multi Attribute Utility (MAU) nodes count:		3
Mean indegree		1.53
Max. indegree:		16
Avg. outcomes count:		2.28
Max. outcomes count:		5

Description:
This is a set of two influence diagrams generated by the DT Tutor IDx. It constructs dynamic decision networks (DDNs) for decision making by computer-based tutors. It decides both the type and the topic of each tutorial action, considering such factors as task progress, the discourse state, and the student's knowledge and focus of attention. A unique DDN is constructed for each problem based on the problem's solution graph.

Reference:
@InProceedings{murray:vanlehn:mostow:04,
author = "Murray, R.C., K. VanLehn, and J. Mostow",
title = "Looking ahead to select tutorial actions: A decision-theoretic approach",
booktitle = "International Journal of Artificial Intelligence in Education,14(3-4)",
year = 2004,
pages = "235--278"
}

ZIP Diabetes

Characteristics:

Nodes count:		413
Mean indegree		1.46
Max. indegree:		2
Avg. outcomes count:		11.33
Max. outcomes count:		21

Description:
A preliminary model for insulin dose adjustment.

Reference:
@InProceedings{andreassen:hovorka:benn:etal:91,
author = "Steen Andreassen and Roman Hovorka and Jonathan Benn and Kristian G. Olesen and Ewart R. Carson",
title = "A Model-based Approach to Insulin Adjustment",
booktitle = "Proceedings of the Third Conference on Artificial Intelligence in Medicine",
year = 1991,
editor = "M. Stefanelli and A. Hasman and M. Fieschi and J. Talmon",
pages = "239--248",
publisher = "Springer-Verlag"
}

ZIP Hailfinder

Characteristics:

Nodes count:		56
Mean indegree		1.18
Max. indegree:		4
Avg. outcomes count:		3.98
Max. outcomes count:		11

Description:
Hailfinder is a normative system that forecasts severe summer hail in northeastern Colorado. It has been generously contributed to the community by Ward Edwards and Bruce Abramson.

Reference:
Abramson, B., J. Brown, W. Edwards, A. Murphy, R. Winkler (1996). Hailfinder: A Bayesian system for forecasting severe weather. International Journal of Forecasting 12(1): 57-72.

ZIP Hepar

Characteristics:

Description:
Hepar II is a Bayesian network model for the diagnosis of liver disorders, generously contributed to the community by Agnieszka Onisko.

Reference:
The primary reference for the Hepar II model is her doctoral dissertation (available at: http://aragorn.pb.bialystok.pl/~aonisko ): Agnieszka Onisko. Probabilistic Causal Models in Medicine: Application to Diagnosis of Liver Disorders. Ph.D. Dissertation, Institute of Biocybernetics and Biomedical Engineering, Polish Academy of Science, Warsaw, March 2003.

ZIP Link

Characteristics:

Nodes count:		724
Mean indegree		1.55
Max. indegree:		3
Avg. outcomes count:		2.53
Max. outcomes count:		4

Description:
A model for the linkage between two genes. One of the genes is the human LQT syndrome (a rare heart disease) and the other is a genetic marker. The model is used to estimate the distance between these two genes.

Reference:
The model originates from Professor Brian Suarez, and has been modified by first Augustine Kong, and later transformed to a Bayesian network by Claus Skaanning Jensen.

@TechReport{jensen:kong:96,
author = {Claus Skaaning Jensen and Augustine Kong},
title = {Blocking {G}ibbs Sampling for Linkage Analysis in Large Pedigrees with Many Loops},
institution = "Department of Computer Science, Aalborg University, Denmark",
year = 1996,
type = {Research Report},
number = {R-96-2048},
address = "Fredrik Bajers Vej 7, DK-9220 Aalborg~\O",
}

ZIP Munin

Characteristics:

Nodes count:		1041
Mean indegree		1.34
Max. indegree:		3
Avg. outcomes count:		5.43
Max. outcomes count:		21

Description:
An expert electromyography assistant.

Reference:
@InCollection{andreassen:jensen:andersen:etal:89,
author = "Andreassen, S. and Jensen, F.V. and Andersen, S.K. and Falck, B. and Kj{\ae}rulff, U. and Woldbye, M. and S{\o}rensen, A.R. and Rosenfalck, A. and Jensen, F.",
title = "{MUNIN} --- An Expert {EMG} Assistant",
booktitle = "Computer-Aided Electromyography and Expert Systems",
publisher = "Elsevier Science Publishers",
address = "Amsterdam",
year = 1989,
editor = "John E. Desmedt",
chapter = 21
}

ZIP Pathfinder

Characteristics:

Nodes count:		109
Mean indegree		1.79
Max. indegree:		5
Avg. outcomes count:		4.11
Max. outcomes count:		63

Description:
An expert system that assists surgical pathologists with the diagnosis of lymph-node diseases.

Reference:
@article{pathfinder,
author={D. E. Heckerman and E. J. Horvitz, and B. N. Nathwani},
title={Toward Normative Expert Systems: Part I The Pathfinder Project},
journal={Methods of Information in Medicine},
year=1992,
volume=31,
pages={90--105}
}

ZIP Win95pts

Characteristics:

Nodes count:		76
Mean indegree		1.47
Max. indegree:		7
Avg. outcomes count:		2
Max. outcomes count:		2

Description:
An expert system for printer troubleshooting in Windows 95.

Reference:
Developed at Microsoft Research and contributed to the community by Jack Breese.