Abstract
Bayesian Networks (BN) is a simple probabilistic graph model built from Bayes probability theory and graph theory. The probability theory is directly related to the data, while the graph theory is directly related to the form of representation to get. BN is a simple Probabilistic Graphical Model (PGM) built from Bayes probability and graph theory. BN can provide simple and solid information about opportunity information. Based on its components, BN consists of Bayesian structure (Bs) and Bayesian parameters (Bp). The constraint base (CB) algorithm is an algorithm that combines two approaches, namely the dependency analysis approach and the search and scoring approach. The purpose of this research is to build a BN structure for incomplete data and to obtain a structure search algorithm that is computationally easy to work and does not require node ordering. The algorithm consists of two phases, namely the first phase is obtained (as part of) the CB algorithm, the result is node ordering. The second phase is designed to study the BN structure from data that has missing values, which is the same as that applied by the Bound and Collapse (BC) algorithm. BN has two algorithms that can work on complete and incomplete databases, namely the hybrid algorithm. Hybrid Algorithm is an algorithm that combines two methods in building structural construction, namely dependency analysis and search and scoring methods. The Hybrid Algorithm can construct structures in the form of graphs and relationships between nodes and display variable probability values based on complete and incomplete database inputs.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have