Complexities and Challenges in the Use of Bayesian Belief Networks: Informing the Design of Causal Influence Models

Abstract

Bayesian belief networks (BNs) are well-suited to capturing vague and uncertain knowledge. However, the capture of this knowledge and associated reasoning from human domain experts often requires specialized knowledge engineers responsible for translating the expert's communications into BN-based models. Across application domains, we have analyzed how these models are constructed, refined, and validated with domain experts. From this analysis, we have identified key user-centered complexities and challenges that we have used to drive the selection of simplifying assumptions. This led us to develop computational techniques and user interface methods that leverage these same assumptions with the goal of improving the efficiency and ease with which expert knowledge can be expressed, verified, validated, and encoded. In this paper, we present the results of our analysis of BN construction, validation, and use. We discuss how these results motivated the design of a simplified version of BNs called Causal Influence Models (CIMs). In addition, we detail how CIMs enable the design and construction of user interface mechanisms that address complexities identified in our analysis.

Get full access to this article

View all access options for this article.

References

Das

(2004). Generating Conditional Probabilities for Bayesian Networks: Easing the Knowledge Acquisition Problem. CoRR, cs.AI/0411034.

Diez

Druzdzel

(2006). Canonical Probabilistic Models for Knowledge Engineering. (Rep. No. Technical Report CISIAD-06–01). Madrid: UNED.

Diez

Galan

S. F.

(2003). An Efficient Factorization for the Noisy MAX. Int'l J. of Intelligent Systems, 18165–177.

Farry

Pfautz

Cox

Bisantz

Roth

(2008). An Experimental Procedure for Evaluating User-Centered Methods for Rapid Bayesian Network Construction. In Proc. of 6th Bayesian Modeling App'ns Workshop.

Guarino

Pfautz

Cox

Roth

(2006). Modeling Human Reasoning about Meta-Information. In Proc. of 4th Bayesian Modeling App'ns Workshop at UAI 1906.

Helander

M. G.

(1987). Design of Visual Displays. In Salvendy

(Ed.), Handbook of Human Factors. NY: Wiley.

Helsper

Van Der Gaag

Feelders

Loeffen

Geenen

Albers

(2005). Bringing Order into Bayesian Network Construction. In Proc. of K-CAP1905.

Henrion

(1989). Some Practical Issues in Constructing Belief Networks. In Kanal

Levitt

Lemmer

(Eds.), Uncertainty in AI 3 (pp. 161–173). North Holland: Elsevier.

Jensen

F. V.

(2001). Bayesian Networks and Decision Graphs. New York: Springer-Verlag.

10.

Koelle

Pfautz

Farry

Cox

Catto

Campolongo

(2006). App'ns of Bayesian Belief Networks in Social Network Analysis. In Proc. of 4th Bayesian Modeling App'ns Workshop at UAI 1906. Cambridge, MA.

11.

Nadkarni

Shenoy

P. P.

(2004). A Causal Mapping Approach to Constructing Bayesian Networks. Decision Support Systems, 38(2), 259–281.

12.

O'hagan

Buck

Daneshkhah

Eiser

Garthwaite

Jenkinson

(2006). Uncertain Judgements: Eliciting Experts' Probabilities. New York: Wiley & Sons.

13.

Pang

Zhang

Wang

(2004). Computerized Tongue Diagnosis Based on Bayesian Networks. IEEE Trans. on Biomedical Eng, 51(10).

14.

Pfautz

Cox

Koelle

Catto

Campolongo

Roth

(2007a). User-Centered Methods for Rapid Creation and Validation of Bayesian Networks. In Proc. of 5th Bayesian App'ns Workshop at UAI 1907. Vancouver.

15.

Pfautz

Fouse

Von Wiegand

Roth

Fichtl

(2006a). An Approach for Multimodal Display Development for UAV Systems. In Proc. of 3rd Ann. Workshop on Human Factors of Unmanned Aerial Vehicles. Mesa, AZ.

16.

Pfautz

Roth

(2006). Using Cognitive Engineering for System Design and Evaluation: A Visualization Aid for Stability and Support Operations. Int'l J. of Industrial Engineering, 36(5), 389–407.

17.

Pfautz

Roth

Powell

Fichtl

Guarino

Carlson

(2006b). Cognitive Complexities Impacting Army Intelligence Analysis. In Proc. of HFES Society 50th Ann. Mtg. San Francisco, California.

18.

Rimey

Brown

(1994). Control of Selective Perception Using Bayes Nets and Decision Theory. Int'l J. of Computer Vision, 12(2–3), 173–207.

19.

Rosen

Smith

(1996). Influence Net Modeling with Causal Strengths. In Proc. of CCRTS.

20.

Roth

E. M.

Stilson

Scott

Whitaker

(2006). Work-Centered Design and Evaluation of a C2 Visualization Aid. In Proc. of HFES Ann. Mtg.

21.

Skaanning

(2000). A Knowledge Acquisition Tool for Bayesian-Network Troubleshooters. In Proc. UAI 1900.

22.

Tabachneck-Schijf

Greenen

(2006). Preventing Knowledge Transfer Errors: Probabilistic Decision Support Systems through the Users' Eyes. In Proc. of 4th Ann. Bayesian Modeling App'ns Workshop at UAI 1906.