Chapter title |
AUC-Maximized Deep Convolutional Neural Fields for Protein Sequence Labeling
|
---|---|
Chapter number | 1 |
Book title |
Machine Learning and Knowledge Discovery in Databases
|
Published in |
Lecture notes in computer science, September 2016
|
DOI | 10.1007/978-3-319-46227-1_1 |
Pubmed ID | |
Book ISBNs |
978-3-31-946226-4, 978-3-31-946227-1
|
Authors |
Sheng Wang, Siqi Sun, Jinbo Xu |
Editors |
Paolo Frasconi, Niels Landwehr, Giuseppe Manco, Jilles Vreeken |
Abstract |
Deep Convolutional Neural Networks (DCNN) has shown excellent performance in a variety of machine learning tasks. This paper presents Deep Convolutional Neural Fields (DeepCNF), an integration of DCNN with Conditional Random Field (CRF), for sequence labeling with an imbalanced label distribution. The widely-used training methods, such as maximum-likelihood and maximum labelwise accuracy, do not work well on imbalanced data. To handle this, we present a new training algorithm called maximum-AUC for DeepCNF. That is, we train DeepCNF by directly maximizing the empirical Area Under the ROC Curve (AUC), which is an unbiased measurement for imbalanced data. To fulfill this, we formulate AUC in a pairwise ranking framework, approximate it by a polynomial function and then apply a gradient-based procedure to optimize it. Our experimental results confirm that maximum-AUC greatly outperforms the other two training methods on 8-state secondary structure prediction and disorder prediction since their label distributions are highly imbalanced and also has similar performance as the other two training methods on solvent accessibility prediction, which has three equally-distributed labels. Furthermore, our experimental results show that our AUC-trained DeepCNF models greatly outperform existing popular predictors of these three tasks. The data and software related to this paper are available at https://github.com/realbigws/DeepCNF_AUC. |
Mendeley readers
Geographical breakdown
Country | Count | As % |
---|---|---|
Israel | 1 | 2% |
United States | 1 | 2% |
Unknown | 55 | 96% |
Demographic breakdown
Readers by professional status | Count | As % |
---|---|---|
Student > Bachelor | 12 | 21% |
Student > Ph. D. Student | 11 | 19% |
Student > Master | 7 | 12% |
Other | 5 | 9% |
Researcher | 5 | 9% |
Other | 8 | 14% |
Unknown | 9 | 16% |
Readers by discipline | Count | As % |
---|---|---|
Computer Science | 18 | 32% |
Biochemistry, Genetics and Molecular Biology | 14 | 25% |
Engineering | 4 | 7% |
Agricultural and Biological Sciences | 2 | 4% |
Chemistry | 2 | 4% |
Other | 6 | 11% |
Unknown | 11 | 19% |