An Algorithm for Generating Concepts Based on Search Space Partition

摘要: Concept lattice, the core data structure in formal concept analysis, has been used widely machine learning, mining and knowledge discovery, information retrieval, etc. The main difficulty with lattice-based system comes from lattice construction itself. This paper proposes a new algorithm called SSPCG (search space partition based concepts generation) on closures search partition. divides into several subspaces accordance criterions prescribed ahead, introduces an efficient scheme to recognize valid ones, which bounds searching just these subspaces. An intermediate is employed judge validity of subspace compute more efficiently. Since recursive independent, parallel version can be directly reached. experimental evaluated compared famous NextClosure proposed by Ganter for random generated data, as well real application data. results show that