Efficient implementation of type inclusion testing is important for data and knowledge base systems employing large hierarchies. The bit vector encoding of a partially ordered set representing a type hierarchy permits constant-time type inclusion testing. Current such methods employ a simple encoding, associating a single gene for each join-irreducible element. We present an algorithm using multiple genes for those elements with many siblings. The new algorithm provides a significant improvement on the encoding size for hierarchies with low multiple inheritance factors. © 2008 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Van Bommel, M., & Wang, P. (2008). Hierarchy encoding with multiple genes. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5181 LNCS, pp. 761–769). https://doi.org/10.1007/978-3-540-85654-2_68
Mendeley helps you to discover research relevant for your work.