A database perspective on knowledge discovery
Communications of the ACM
Fast discovery of association rules
Advances in knowledge discovery and data mining
A relational model of data for large shared data banks
Communications of the ACM
Frequent Closures as a Concise Representation for Binary Data Mining
PADKK '00 Proceedings of the 4th Pacific-Asia Conference on Knowledge Discovery and Data Mining, Current Issues and New Applications
A perspective on inductive databases
ACM SIGKDD Explorations Newsletter
Interestingness of frequent itemsets using Bayesian networks as background knowledge
Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining
Elements of Information Theory (Wiley Series in Telecommunications and Signal Processing)
Elements of Information Theory (Wiley Series in Telecommunications and Signal Processing)
On data mining, compression, and Kolmogorov complexity
Data Mining and Knowledge Discovery
Preserving Privacy through Data Generation
ICDM '07 Proceedings of the 2007 Seventh IEEE International Conference on Data Mining
Learning Bayesian network parameters under order constraints
International Journal of Approximate Reasoning
Compression picks item sets that matter
PKDD'06 Proceedings of the 10th European conference on Principle and Practice of Knowledge Discovery in Databases
| Hi-index | 0.00 |
Inductive databases are databases in which models and patterns are first class citizens. Having models and patterns in the database raises the question: do the models and patterns that are stored help in computing new models and patterns? For example, let C be a classifier on database DB and let Q be a query. Does knowing C speed up the induction of a new classifier on the result of Q ? In this paper we answer this problem positively for the code tables induced by our Krimp algorithm. More in particular, assume we have the code tables for all tables in the database. Then we can approximate the code table induced by Krimp on the result of a query, using only these global code tables as candidates. That is, we do not have to mine for frequent item sets on the query result.