Exploring constraints to efficiently mine emerging patterns from large high-dimensional datasets

@inproceedings{Zhang2000ExploringCT,
  title={Exploring constraints to efficiently mine emerging patterns from large high-dimensional datasets},
  author={Xiuzhen Zhang and Guozhu Dong and Kotagiri Ramamohanarao},
  booktitle={KDD},
  year={2000}
}
Emerging patterns (EPs) were proposed re ently to apture hanges or di eren es between datasets: an EP is a multivariate feature whose support in reases sharply from a ba kground dataset to a target dataset, and the support ratio is alled its growth rate. Interesting long EPs often have low support; mining su h EPs from high-dimensional datasets is a great hallenge due to the ombinatorial explosion of the number of andidates. We propose a Constraint-based EP Miner, ConsEPMiner, that utilizes two… CONTINUE READING

From This Paper

Figures, tables, and topics from this paper.

References

Publications referenced by this paper.

Constraint-Based Rule Mining in Large, Dense Databases

Data Mining and Knowledge Discovery • 1999
View 6 Excerpts
Highly Influenced

Similar Papers

Loading similar papers…