Semi-supervised Convolutional Neural Networks for Text Categorization via Region Embedding

Abstract

This paper presents a new semi-supervised framework with convolutional neural networks (CNNs) for text categorization. Unlike the previous approaches that rely on word embeddings, our method learns embeddings of small text regions from unlabeled data for integration into a supervised CNN. The proposed scheme for embedding learning is based on the idea of two-view semi-supervised learning, which is intended to be useful for the task of interest even though the training is done on unlabeled data. Our models achieve better results than previous approaches on sentiment classification and topic classification tasks.

arXiv commentv1 has a different title, and the results there are obsolete. The current version is to appear in NIPS 2015

References in corpus (3)

• Efficient Estimation of Word Representations in Vector Space
• Improving neural networks by preventing co-adaptation of feature detectors
• Ensemble of Generative and Discriminative Techniques for Sentiment Analysis of Movie Reviews

Cited by in corpus (24)

• Representation learning for very short texts using weighted word embedding aggregation
• Adversarial Training Methods for Semi-Supervised Text Classification
• Supervised and Semi-Supervised Text Categorization using LSTM for Region Embeddings
• A Survey on Text Classification: From Shallow to Deep Learning
• Transformation Networks for Target-Oriented Sentiment Classification
• Deep Learning-based Sentiment Classification: A Comparative Survey
• Convolutional Neural Networks for Text Categorization: Shallow Word-level vs. Deep Character-level
• Interpretable Adversarial Perturbation in Input Embedding Space for Text
• Kronecker Attention Networks
• Rationale-Augmented Convolutional Neural Networks for Text Classification
• Chat Detection in an Intelligent Assistant: Combining Task-oriented and Non-task-oriented Spoken Dialogue Systems
• RETRIEVE: Coreset Selection for Efficient and Robust Semi-Supervised Learning
• Learning Norms from Stories: A Prior for Value Aligned Agents
• iCapsNets: Towards Interpretable Capsule Networks for Text Classification
• Can Ground Truth Label Propagation from Video help Semantic Segmentation?
• Sparse Autoencoder for Unsupervised Nucleus Detection and Representation in Histopathology Images
• Reading Comprehension using Entity-based Memory Network
• Multiple Instance Learning Networks for Fine-Grained Sentiment Analysis
• Character-Based Text Classification using Top Down Semantic Model for Sentence Representation
• Investigating the Working of Text Classifiers
• Mixture of Expert/Imitator Networks: Scalable Semi-supervised Learning Framework
• Image-based Natural Language Understanding Using 2D Convolutional Neural Networks
• Strategies for Language Identification in Code-Mixed Low Resource Languages
• Patient Risk Assessment and Warning Symptom Detection Using Deep Attention-Based Neural Networks