A multi-source Tri-Training transfer learning algorithm is proposed by integrating transfer learning and semi-supervised learning. First, multiple weak classifiers are respectively trained by using both weighted source and target training samples. Then, based on the idea of co-training, each target testing sample is labeled by using trained weak classifiers and the sample with the same label is selected as the high-confidence sample to be added into the target training sample set. Finally, we can obtain a target domain classifier based on the updated target training samples. The above steps are iterated till the high-confidence samples selected at two successive iterations become the same. At each iteration, source training samples are tested by using the target domain classifier and the samples tested as correct continue with training, while the weights of samples tested as incorrect are lowered. Experimental results on text classification dataset have proven the effectiveness and superiority of the proposed algorithm.
Yuhu CHENG
China University of Mining and Technology
Xuesong WANG
China University of Mining and Technology
Ge CAO
China University of Mining and Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yuhu CHENG, Xuesong WANG, Ge CAO, "Multi-Source Tri-Training Transfer Learning" in IEICE TRANSACTIONS on Information,
vol. E97-D, no. 6, pp. 1668-1672, June 2014, doi: 10.1587/transinf.E97.D.1668.
Abstract: A multi-source Tri-Training transfer learning algorithm is proposed by integrating transfer learning and semi-supervised learning. First, multiple weak classifiers are respectively trained by using both weighted source and target training samples. Then, based on the idea of co-training, each target testing sample is labeled by using trained weak classifiers and the sample with the same label is selected as the high-confidence sample to be added into the target training sample set. Finally, we can obtain a target domain classifier based on the updated target training samples. The above steps are iterated till the high-confidence samples selected at two successive iterations become the same. At each iteration, source training samples are tested by using the target domain classifier and the samples tested as correct continue with training, while the weights of samples tested as incorrect are lowered. Experimental results on text classification dataset have proven the effectiveness and superiority of the proposed algorithm.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E97.D.1668/_p
Copy
@ARTICLE{e97-d_6_1668,
author={Yuhu CHENG, Xuesong WANG, Ge CAO, },
journal={IEICE TRANSACTIONS on Information},
title={Multi-Source Tri-Training Transfer Learning},
year={2014},
volume={E97-D},
number={6},
pages={1668-1672},
abstract={A multi-source Tri-Training transfer learning algorithm is proposed by integrating transfer learning and semi-supervised learning. First, multiple weak classifiers are respectively trained by using both weighted source and target training samples. Then, based on the idea of co-training, each target testing sample is labeled by using trained weak classifiers and the sample with the same label is selected as the high-confidence sample to be added into the target training sample set. Finally, we can obtain a target domain classifier based on the updated target training samples. The above steps are iterated till the high-confidence samples selected at two successive iterations become the same. At each iteration, source training samples are tested by using the target domain classifier and the samples tested as correct continue with training, while the weights of samples tested as incorrect are lowered. Experimental results on text classification dataset have proven the effectiveness and superiority of the proposed algorithm.},
keywords={},
doi={10.1587/transinf.E97.D.1668},
ISSN={1745-1361},
month={June},}
Copy
TY - JOUR
TI - Multi-Source Tri-Training Transfer Learning
T2 - IEICE TRANSACTIONS on Information
SP - 1668
EP - 1672
AU - Yuhu CHENG
AU - Xuesong WANG
AU - Ge CAO
PY - 2014
DO - 10.1587/transinf.E97.D.1668
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E97-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2014
AB - A multi-source Tri-Training transfer learning algorithm is proposed by integrating transfer learning and semi-supervised learning. First, multiple weak classifiers are respectively trained by using both weighted source and target training samples. Then, based on the idea of co-training, each target testing sample is labeled by using trained weak classifiers and the sample with the same label is selected as the high-confidence sample to be added into the target training sample set. Finally, we can obtain a target domain classifier based on the updated target training samples. The above steps are iterated till the high-confidence samples selected at two successive iterations become the same. At each iteration, source training samples are tested by using the target domain classifier and the samples tested as correct continue with training, while the weights of samples tested as incorrect are lowered. Experimental results on text classification dataset have proven the effectiveness and superiority of the proposed algorithm.
ER -