Summit-Training: A hybrid Semi-Supervised technique and its application to classification tasks

详细信息    查看全文

• 作者：L. Tencer ; ^{lukas.tencer@gmail.com} ; ; M. Reznakova ^{marta.reznakova@gmail.com} ; M. Cheriet ^{mohamed.cheriet@etsmtl.ca}
• 关键词：Hybrid methods ; Semi-supervised learning ; Classification ; Summit-Training ; Query-by-Committee ; Unlabeled data
• 刊名：Applied Soft Computing
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：50
• 期：Complete
• 页码：1-20
• 全文大小：5733 K
• 卷排序：50

文摘

We propose a novel hybrid Semi-Supervised technique (Summit-Training) for classification tasks. Summit-Training combines elements of Semi-Supervised training and Active Learning. We evaluated it on six datasets and applied to 13 classifiers resulting in 78 test cases. It improves performance of all classifiers, outperforms other baseline Semi-Supervised techniques. We use Summit-Training in gesture recognition which is deployed in real-world HCI system.