Title

Gated bayesian networks for medical decision support systems

Degree

Master of Science in Computer Science

Faculty / School

Faculty of Computer Sciences (FCS)

Department

Department of Computer Science

Date of Award

Summer 2018

Advisor/Supervisor

Dr. Muhammad Saeed

Committee Member 1

Dr. Muhammad Saeed, Supervisor, Department of Computer Science, Institute of Business Administration (IBA), Karachi.

Project Type

MSCS Survey Report

Access Type

Restricted Access

Keywords

Machine learning, Bayesian statistical decision theory--Data processing, Neural networks (Computer science).

Subjects

Computer Science

Abstract / Summary

There are various problems where we need to switch between different phases with respect to some criteria. This study will be focused on the new probabilistic graphical model named "Gated Bayesian Network". A Gated Bayesian network (GBN) is a set of various Bayesian networks which are joint with the help of nodes called "gates". With the help of predefined rules, it will be decided that which gate to be triggered and switch into a particular Bayesian network. It is like a scenario where a situation changes in back and forth motion with respect to the predefmed rules on the current situation. Various real-life scenarios can be modeled using this approach. This study is specifically highlighting the application of gated Bayesian network in medical sciences where it is helpful in the construction of automated medical decision support system. Glimpse ofGBN in finance domain has also discussed in detailed. In this study, we will discuss this new formalism, its execution in, implementation results and comparison with other strategies.

Pages

1

This document is currently not available here.

The full text of this document is only accessible to authorized users.

Share

COinS