A cost-reducing partial labeling estimator in text classification problem
Machine Learning
2019-06-11 v1 Information Retrieval
Machine Learning
Abstract
We propose a new approach to address the text classification problems when learning with partial labels is beneficial. Instead of offering each training sample a set of candidate labels, we assign negative-oriented labels to the ambiguous training examples if they are unlikely fall into certain classes. We construct our new maximum likelihood estimators with self-correction property, and prove that under some conditions, our estimators converge faster. Also we discuss the advantages of applying one of our estimator to a fully supervised learning problem. The proposed method has potential applicability in many areas, such as crowdsourcing, natural language processing and medical image analysis.
Cite
@article{arxiv.1906.03768,
title = {A cost-reducing partial labeling estimator in text classification problem},
author = {Jiangning Chen and Zhibo Dai and Juntao Duan and Qianli Hu and Ruilin Li and Heinrich Matzinger and Ionel Popescu and Haoyan Zhai},
journal= {arXiv preprint arXiv:1906.03768},
year = {2019}
}