Efficient implementation of lattice operations
ACM Transactions on Programming Languages and Systems (TOPLAS)
Mining knowledge at multiple concept levels
CIKM '95 Proceedings of the fourth international conference on Information and knowledge management
Data mining: concepts and techniques
Data mining: concepts and techniques
Formal concept mining: a statistic-based approach for pertinent concept lattice construction
ASIAN'04 Proceedings of the 9th Asian Computing Science conference on Advances in Computer Science: dedicated to Jean-Louis Lassez on the Occasion of His 5th Cycle Birthday
| Hi-index | 0.00 |
With the rapid growth in size and number of available databases, the manipulation of large concept hierarchies that cannot be fit in main memory becomes more and more frequent. Several representations of concept hierarchies are possible, for example tree, lattice, table, linked list, etc. In this paper, we propose an efficient implementation technique to manipulate large concept hierarchies. We use a lattice data structure to represent concept hierarchies and encode such a lattice into a boolean transitive closure matrix. A set of lattice operators are defined and implemented as abstract data types on the top of an object-relational database management system, and are used to perform generalization and specialization operations. We show the efficiency of the lattice operators to perform generalization and specialization in large concept hierarchies and compare their performance with the START WITH and CONNECT BY clauses of SQL.