This article is cited in 2 papers

Algebraic Bayesian Network Tertiary Polystructure Elements Synthesis Algorithms

A. A. Fil'chenkov

St. Petersburg Institute for Informatics and Automation of RAS

Abstract: The role of algebraic Bayesian network (ABN) polystructure has increased significantly. It had been originally introduced as an auxiliary object for the secondary structure synthesis, but then the tertiary polystructure has found its use in the analysis of secondary structure cyclicity without its direct synthesis. Now the tertiary polystructure is expected to be used for global inference in ABN. The goal of the work is the selection (and subsequent systematization and complicity estimation) the existing algorithms for tertiary polystructure elements synthesis. The existing algorithms for tertiary polystructure elements synthesis are overviewed and the algorithm complicity is estimated in the paper. Four algorithms for synthesizing the empty graph over useful cliques subsets and two algorithms for synthesizing the parent graph over a stereoclique set are presented.

Keywords: algebraic Bayesian networks, tertiary polystructure, machine learning, probabilistic graphical models, global structure.

UDC: 004.8

Received: 02.08.2011
Accepted: 29.09.2011