The evolutionary learning method of Bayesian network structure based on expert knowledge

Jiahui Wang; Lei Zhao; Yan Yan; Jia Hao

doi:10.1117/12.2640077

9 May 2022 The evolutionary learning method of Bayesian network structure based on expert knowledge

Jiahui Wang, Lei Zhao, Yan Yan, Jia Hao

Proceedings Volume 12252, International Conference on Biometrics, Microelectronic Sensors, and Artificial Intelligence (BMSAI); 122520P (2022) https://doi.org/10.1117/12.2640077
Event: International Conference on Biometrics, Microelectronic Sensors, and Artificial Intelligence (BMSAI), 2022, Guangzhou, China

Abstract

Bayesian Network, which has been widely used for its significant advantages in causal inference, has great potential in product design. However, it is difficult to learn a reasonable network structure facing the problem with small data during the design process. In order to solve this problem, this paper introduces expert knowledge and Genetic Algorithm based on matrix coding into the process of Bayesian Network structure learning. Integrating the expert knowledge and data is a decent way to make up for the problem of insufficient data, and Genetic Algorithm is used to improve traditional structure learning algorithms, so as to obtain a more suitable structure conforming to the knowledge and data characteristics. The solutions illustrate that the Genetic Algorithm has some advantages compared with traditional structure learning method, and the use of expert knowledge can improve the rationality of the learned structure.

Citation Download Citation

Jiahui Wang, Lei Zhao, Yan Yan, and Jia Hao "The evolutionary learning method of Bayesian network structure based on expert knowledge", Proc. SPIE 12252, International Conference on Biometrics, Microelectronic Sensors, and Artificial Intelligence (BMSAI), 122520P (9 May 2022); https://doi.org/10.1117/12.2640077

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available