AdaSampling for Positive-Unlabeled and Label Noise Learning With Bioinformatics Applications
- 2 April 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 49 (5) , 1932-1943
Abstract
Class labels are required for supervised learning but may be corrupted or missing in various applications. In binary classification, for example, when only a subset of positive instances is labeled whereas the remaining are unlabeled, positive-unlabeled (PU) learning is required to model from both positive and unlabeled data. Similarly, when class labels are corrupted by mislabeled instances, methods are needed for learning in the presence of class label noise (LN). Here we propose adaptive sampling (AdaSampling), a framework for both PU learning and learning with class LN. By iteratively estimating the class mislabeling probability with an adaptive sampling procedure, the proposed method progressively reduces the risk of selecting mislabeled instances for model training and subsequently constructs highly generalizable models even when a large proportion of mislabeled instances is present in the data. We demonstrate the utilities of proposed methods using simulation and benchmark data, and compare them to alternative approaches that are commonly used for PU learning and/or learning with LN. We then introduce two novel bioinformatics applications where AdaSampling is used to: 1) identify kinase-substrates from mass spectrometry-based phosphoproteomics data and 2) predict transcription factor target genes by integrating various next-generation sequencing data.Keywords
Funding Information
- Australian Research Council (DE170100759)
- Australian Research Council (DP170100654)
- Australian Research Council (DP170100654)
- Australian Research Council (DP130104591)
- Australian Research Council (DP170100654)
- National Health and Medical Research Council (1111338)
This publication has 55 references indexed in Scilit:
- Dynamic Adipocyte Phosphoproteome Reveals that Akt Directly Regulates mTORC2Cell Metabolism, 2013
- Positive-unlabeled learning for disease gene identificationBioinformatics, 2012
- A study of the effect of different types of noise on the precision of supervised learning techniquesArtificial Intelligence Review, 2010
- Robust supervised classification with mixture models: Learning from data with uncertain labelsPattern Recognition, 2009
- Learning from partially supervised data using mixture models and belief functionsPattern Recognition, 2009
- Integration of External Signaling Pathways with the Core Transcriptional Network in Embryonic Stem CellsCell, 2008
- Learning from positive and unlabeled examplesTheoretical Computer Science, 2005
- Class Noise vs. Attribute Noise: A Quantitative StudyArtificial Intelligence Review, 2004
- Additive logistic regression: a statistical view of boosting (With discussion and a rejoinder by the authors)The Annals of Statistics, 2000
- Wrappers for feature subset selectionArtificial Intelligence, 1997