Titlebook: Advanced Data Mining and Applications; 5th International Co Ronghuai Huang,Qiang Yang,Xue Li Conference proceedings 2009 Springer-Verlag Be

豎琴

偉大

https://doi.org/10.1007/978-3-662-02038-8lts show that the proposed methods are effective. Several class contrast functions are discovered from the real world datasets. Some potential works on class contrast function mining are discussed based on the experimental results.

無關(guān)緊要

Clinch

https://doi.org/10.1007/978-3-662-02038-8ng to frequently collect sufficient labeled data for training. Experimental results demonstrate that our approach achieves 91.66% high detection rate with less than 2.96% false positive rate, meanwhile it significantly reduces the labeling cost.