Multilabel Text Classification With Incomplete Labels: A Safe Generative Model With Label Manifold Regularization and Confidence Constraint
- 7 September 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE MultiMedia
- Vol. 27 (4) , 38-47
Abstract
In multi-label text classification, the label information of training instances may be incomplete due to various factors, such as expensively manual labeling, errors etc. Hence imposing big challenges to training classifiers. To tackle the problem of the incomplete labels, we proposed to exploit label correlations, and further jointly leverage manifold regularization and label confidence constraint. On one hand, the label manifold regularization preserves the local label structure to handle the high-level incomplete labels. On the other hand, the label confidence constraint, following the likelihood theory, avoids over-estimating the weights of negative labels, leading to a safer inference. Upon the ideas, we propose a novel generative model, namely Bayesian Model with Label manifold Regularization and label Confidence constraint (BM-LRC), solved by generalized expectation maximization. Empirical results on popular benchmark datasets show that BM-LRC achieves competitive performance, especially when the available labels are scarce.Keywords
Funding Information
- National Natural Science Foundation of China (61602204)
This publication has 11 references indexed in Scilit:
- RMoR-Aion: Robust Multioutput Regression by Simultaneously Alleviating Input and Output NoisesIEEE Transactions on Neural Networks and Learning Systems, 2020
- Multi-label Learning with Missing Labels Using Mixed Dependency GraphsInternational Journal of Computer Vision, 2018
- Extreme Multi-label Loss Functions for Recommendation, Tagging, Ranking & Other Missing Label Applications2016
- Robust Extreme Multi-label Learning2016
- ML-MG: Multi-label Learning with Missing Labels Using a Mixed Graph2015
- Binary Codes Embedding for Fast Image Tagging with Incomplete Labels2014
- A Review on Multi-Label Learning AlgorithmsIEEE Transactions on Knowledge and Data Engineering, 2013
- Statistical topic models for multi-label document classificationMachine Learning, 2011
- Modeling hidden topics on document manifold2008
- Pachinko allocation2006