Title :
Research of classification intersection hierarchical granularity algorithm based on concept lattice
Author :
Liu, Baoxiang ; Feng, Li
Author_Institution :
Coll. of Sci., Hebei Polytech. Univ., Tangshan, China
Abstract :
Both of the lattice structure of the concept lattice and hierarchical models of the quotient space can be used to layering, and can be obtained figures under certain conditions. Because of the close links between the above, this paper uses the basic method of the lattice structure, upgrades the direct intersection concept at first, then makes up the additional points for this layer, and searches the equivalence classes with the quotient attribute functions, ensures the integrity of the domain in each layer. On this basis, a new algorithm is proposed which integrates hierarchical models and lattice structures to achieve the layering of classification intersecting situation. Example shows the method feasible.
Keywords :
equivalence classes; lattice theory; pattern classification; classification intersection hierarchical granularity algorithm; concept lattice; direct intersection concept; equivalence classes; hierarchical models; lattice structure; quotient attribute functions; quotient space; Algorithm design and analysis; Analytical models; Classification algorithms; Classification tree analysis; Computational modeling; Lattices; Search problems; concept lattice; granularity; hierarchichy; lattice structure; the quotient attribute function;
Conference_Titel :
Natural Computation (ICNC), 2010 Sixth International Conference on
Conference_Location :
Yantai, Shandong
Print_ISBN :
978-1-4244-5958-2
DOI :
10.1109/ICNC.2010.5583134