Learning with Noisy Labels for Sentence-level Sentiment Classification

Deep neural networks (DNNs) can fit (or even over-fit) the training data very well. If a DNN model is trained using data with noisy labels and tested on data with clean labels, the model may perform poorly. This paper studies the problem of learning with noisy labels for sentence-level sentiment classification. We propose a novel DNN model called NetAb (as shorthand for convolutional neural Networks with Ab-networks) to handle noisy labels during training. NetAb consists of two convolutional neural networks, one with a noise transition layer for dealing with the input noisy labels and the other for predicting ‘clean’ labels. We train the two networks using their respective loss functions in a mutual reinforcement manner. Experimental results demonstrate the effectiveness of the proposed model.

[1]  Shuai Wang,et al.  Deep learning for sentiment analysis: A survey , 2018, WIREs Data Mining Knowl. Discov..

[2]  James Bailey,et al.  Dimensionality-Driven Learning with Noisy Labels , 2018, ICML.

[3]  Michael Gamon,et al.  Sentiment classification on customer feedback data: noisy data, large feature vectors, and the role of linguistic analysis , 2004, COLING.

[4]  Yi Zheng,et al.  Weakly-Supervised Deep Learning for Customer Review Sentiment Classification , 2016, IJCAI.

[5]  Nagarajan Natarajan,et al.  Learning with Noisy Labels , 2013, NIPS.

[6]  Lu Wang,et al.  Risk Minimization in the Presence of Label Noise , 2016, AAAI.

[7]  Li Fei-Fei,et al.  MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labels , 2017, ICML.

[8]  D. Angluin,et al.  Learning From Noisy Examples , 1988, Machine Learning.

[9]  Jacob Goldberger,et al.  Training deep neural-networks based on unreliable labels , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[10]  Dacheng Tao,et al.  Classification with Noisy Labels by Importance Reweighting , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Jun Zhao,et al.  Adding Redundant Features for CRFs-based Sentence Sentiment Classification , 2008, EMNLP.

[12]  Mert R. Sabuncu,et al.  Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels , 2018, NeurIPS.

[13]  Nagarajan Natarajan,et al.  Cost-Sensitive Learning with Noisy Labels , 2017, J. Mach. Learn. Res..

[14]  Alok N. Choudhary,et al.  Sentiment Analysis of Conditional Sentences , 2009, EMNLP.

[15]  Bing Liu,et al.  Mining and summarizing customer reviews , 2004, KDD.

[16]  Yoon Kim,et al.  Convolutional Neural Networks for Sentence Classification , 2014, EMNLP.

[17]  Ming Zhou,et al.  A Joint Segmentation and Classification Framework for Sentence Level Sentiment Classification , 2015, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[18]  Yale Song,et al.  Learning from Noisy Labels with Distillation , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[19]  Bo Pang,et al.  Seeing Stars: Exploiting Class Relationships for Sentiment Categorization with Respect to Rating Scales , 2005, ACL.

[20]  Yoshua Bengio,et al.  A Closer Look at Memorization in Deep Networks , 2017, ICML.

[21]  Matthew S. Nokleby,et al.  Learning Deep Networks from Noisy Labels with Dropout Regularization , 2016, 2016 IEEE 16th International Conference on Data Mining (ICDM).

[22]  Jun Yan,et al.  Sentence-level Sentiment Classification with Weak Supervision , 2017, SIGIR.

[23]  Jeffrey Pennington,et al.  GloVe: Global Vectors for Word Representation , 2014, EMNLP.

[24]  Richard Nock,et al.  Making Deep Neural Networks Robust to Label Noise: A Loss Correction Approach , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[25]  Lei Zhang,et al.  Sentiment Analysis and Opinion Mining , 2017, Encyclopedia of Machine Learning and Data Mining.

[26]  Yanghui Rao,et al.  Learning from Multi-annotator Data , 2019, ACM Trans. Inf. Syst..

[27]  Junlan Feng,et al.  Robust Sentiment Detection on Twitter from Biased and Noisy Data , 2010, COLING.

[28]  Xiaoyan Zhu,et al.  Sentiment Analysis by Capsules , 2018, WWW.

[29]  Christopher D. Manning,et al.  Baselines and Bigrams: Simple, Good Sentiment and Topic Classification , 2012, ACL.

[30]  Aritra Ghosh,et al.  Robust Loss Functions under Label Noise for Deep Neural Networks , 2017, AAAI.

[31]  Navneet Kaur,et al.  Opinion mining and sentiment analysis , 2016, 2016 3rd International Conference on Computing for Sustainable Global Development (INDIACom).

[32]  Abhinav Gupta,et al.  Learning from Noisy Large-Scale Datasets with Minimal Supervision , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[33]  Shai Shalev-Shwartz,et al.  Decoupling "when to update" from "how to update" , 2017, NIPS.

[34]  Samy Bengio,et al.  Understanding deep learning requires rethinking generalization , 2016, ICLR.

[35]  Xingrui Yu,et al.  Co-teaching: Robust training of deep neural networks with extremely noisy labels , 2018, NeurIPS.

[36]  Ivor W. Tsang,et al.  Masking: A New Perspective of Noisy Supervision , 2018, NeurIPS.

[37]  Xiaogang Wang,et al.  Learning from massive noisy labeled data for image classification , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[38]  Oscar Täckström,et al.  Semi-supervised latent variable models for sentence-level sentiment analysis , 2011, ACL.

[39]  Claire Cardie,et al.  Context-aware Learning for Sentence-level Sentiment Analysis with Posterior Regularization , 2014, ACL.

[40]  Minyi Guo,et al.  Emoticon Smoothed Language Models for Twitter Sentiment Analysis , 2012, AAAI.

[41]  Arash Vahdat,et al.  Toward Robustness against Label Noise in Training Deep Discriminative Neural Networks , 2017, NIPS.

[42]  Bin Yang,et al.  Learning to Reweight Examples for Robust Deep Learning , 2018, ICML.

[43]  Joan Bruna,et al.  Training Convolutional Networks with Noisy Labels , 2014, ICLR 2014.

[44]  Jacob Goldberger,et al.  Training deep neural-networks using a noise adaptation layer , 2016, ICLR.

[45]  Yanghui Rao,et al.  Sentiment and emotion classification over noisy labels , 2016, Knowl. Based Syst..

[46]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[47]  Dumitru Erhan,et al.  Training Deep Neural Networks on Noisy Labels with Bootstrapping , 2014, ICLR.

[48]  Geoffrey E. Hinton,et al.  Learning to Label Aerial Images from Noisy Data , 2012, ICML.

[49]  Lillian Lee,et al.  Opinion Mining and Sentiment Analysis , 2008, Found. Trends Inf. Retr..

[50]  Cheng Soon Ong,et al.  Learning from Corrupted Binary Labels via Class-Probability Estimation , 2015, ICML.