Active Learning Based on diversity maximization
In many practical data mining applications,unlabeled training examples arc readily available but labeled ones are fairly expensive to obtain.Therefore,as one type of the paradigms for addressing the problem of combining labeled and unlabeled data to boost the performance,active learning has attracted much attention.In this paper,we propose a new active learning approach based on diversity maximization.Different from the well-known co-testing algorithm,our method does not require two different views.The comparative studies with other active learning methods demonstrate the effectiveness of tbe proposed approach.
machine learning active learning classification diversity
Yongcheng Wu
Computing school Jingchu University of Technology Jingmen, Hubei, 448000, China
国际会议
太原
英文
822-825
2013-04-06（万方平台首次上网日期，不代表论文的发表时间）