Degree Type

Creative Component

Semester of Graduation

Spring 2018


Computer Science

First Major Professor

Jin Tian


Master of Science (MS)


Computer Science


Bayesian networks are being used in various domains, such as data mining, diagnosis, bioinformatics/computational biology, etc. One problem associated with Bayesian networks is to learn their structures from training data. In this paper, we introduce a new approach to structural learning of Bayesian networks, based on hierarchical clustering. We learn the network in hierarchical stages, learning over a subset of the random variables at each stage. Experiments show that this approach learns Bayesian networks faster as compared to curriculum-based learning methods. We show a comparison of our networks with curriculum based learned Bayesian networks over different evaluation metrics as well. Also, performance of hierarchical clustering vs an existing ordering-based algorithm is observed.

Copyright Owner

Nikhita Sharma

File Format