Character-level Convolutional Networks for Text Classification

Abstract

This article offers an empirical exploration on the use of character-level convolutional networks (ConvNets) for text classification. We constructed several large-scale datasets to show that character-level convolutional networks could achieve state-of-the-art or competitive results. Comparisons are offered against traditional models such as bag of words, n-grams and their TFIDF variants, and deep learning models such as word-based ConvNets and recurrent neural networks.

An early version of this work entitled "Text Understanding from Scratch" was posted in Feb 2015 as arXiv:1502.01710. The present paper has considerably more experimental results and a rewritten introduction, Advances in Neural Information Processing Systems 28 (NIPS 2015)

References in corpus (4)

• LSTM: A Search Space Odyssey
• Improving neural networks by preventing co-adaptation of feature detectors
• Natural Language Processing (almost) from Scratch
• On the difficulty of training Recurrent Neural Networks

Cited by in corpus (156)

• Multitask Prompted Training Enables Zero-Shot Task Generalization
• HDLTex: Hierarchical Deep Learning for Text Classification
• A Survey on Data Augmentation for Text Classification
• FedALA: Adaptive Local Aggregation for Personalized Federated Learning
• Deep Text Classification Can be Fooled
• Data Augmentation Approaches in Natural Language Processing: A Survey
• Unleashing the potential of prompt engineering for large language models
• TextBugger: Generating Adversarial Text Against Real-world Applications
• Split Computing and Early Exiting for Deep Learning Applications: Survey and Research Challenges
• Efficiently Trainable Text-to-Speech System Based on Deep Convolutional Networks with Guided Attention
• "What is Relevant in a Text Document?": An Interpretable Machine Learning Approach
• An Analysis of Hierarchical Text Classification Using Word Embeddings
• Natural Language Inference over Interaction Space
• DKN: Deep Knowledge-Aware Network for News Recommendation
• TweepFake: about Detecting Deepfake Tweets
• Weakly-Supervised Neural Text Classification
• Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications
• Learning the PE Header, Malware Detection with Minimal Domain Knowledge
• CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation
• The value of text for small business default prediction: A deep learning approach
• Supervised and Unsupervised Neural Approaches to Text Readability
• LAMOL: LAnguage MOdeling for Lifelong Language Learning
• A Survey on Data-driven Software Vulnerability Assessment and Prioritization
• Entailment as Few-Shot Learner
• FedCP: Separating Feature Information for Personalized Federated Learning via Conditional Policy
• Generative Data Augmentation for Commonsense Reasoning
• Predicting Network Controllability Robustness: A Convolutional Neural Network Approach
• Revisiting LSTM Networks for Semi-Supervised Text Classification via Mixed Objective Function
• Graph Neural Networks for Natural Language Processing: A Survey
• Fine-grained Visual-textual Representation Learning
• Text Classification using Capsules
• Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventions
• Analysis of Railway Accidents' Narratives Using Deep Learning
• Deep Learning based Emotion Recognition System Using Speech Features and Transcriptions
• AdaGNN: Graph Neural Networks with Adaptive Frequency Response Filter
• Deep Learning-based Sentiment Classification: A Comparative Survey
• Facilitating Knowledge Sharing from Domain Experts to Data Scientists for Building NLP Models
• Transfer Learning and Distant Supervision for Multilingual Transformer Models: A Study on African Languages
• Towards Zero-Label Language Learning
• Denoising Multi-Source Weak Supervision for Neural Text Classification
• Benchmark and application of unsupervised classification approaches for univariate data
• Mask-guided BERT for Few Shot Text Classification
• Natural Language Understanding with Privacy-Preserving BERT
• CoAnnotating: Uncertainty-Guided Work Allocation between Human and Large Language Models for Data Annotation
• Natural Language Understanding for Argumentative Dialogue Systems in the Opinion Building Domain
• Tree-structured multi-stage principal component analysis (TMPCA): theory and applications
• Image Captioning with Very Scarce Supervised Data: Adversarial Semi-Supervised Learning Approach
• A Comparison of Machine Learning Algorithms for the Surveillance of Autism Spectrum Disorder
• Spinning Language Models: Risks of Propaganda-As-A-Service and Countermeasures
• Keyphrase Extraction from Disaster-related Tweets
• Are Pre-trained Convolutions Better than Pre-trained Transformers?
• A Fast Unified Model for Parsing and Sentence Understanding
• Deep Character-Level Click-Through Rate Prediction for Sponsored Search
• Attention Meets Perturbations: Robust and Interpretable Attention with Adversarial Training
• Weakly Supervised Multi-Task Representation Learning for Human Activity Analysis Using Wearables
• Federated Few-Shot Learning for Mobile NLP
• Artificial intelligence for topic modelling in Hindu philosophy: mapping themes between the Upanishads and the Bhagavad Gita
• Who Needs Words? Lexicon-Free Speech Recognition
• "That Is a Suspicious Reaction!": Interpreting Logits Variation to Detect NLP Adversarial Attacks
• Enhanced word embeddings using multi-semantic representation through lexical chains
• Feature engineering vs. deep learning for paper section identification: Toward applications in Chinese medical literature
• Interpreting Deep Learning Models in Natural Language Processing: A Review
• Opinion Conflicts: An Effective Route to Detect Incivility in Twitter
• Kurdish Handwritten Character Recognition using Deep Learning Techniques
• SIMILAR: Submodular Information Measures Based Active Learning In Realistic Scenarios
• Improving the Adversarial Robustness of NLP Models by Information Bottleneck
• Fake News Detection: a comparison between available Deep Learning techniques in vector space
• Near-lossless Binarization of Word Embeddings
• Economic Recommender Systems -- A Systematic Review
• Convolutional Neural Network: Text Classification Model for Open Domain Question Answering System
• DeTiME: Diffusion-Enhanced Topic Modeling using Encoder-decoder based LLM
• Analysis and Optimization of fastText Linear Text Classifier
• Social Bots for Online Public Health Interventions
• Modeling citation worthiness by using attention-based bidirectional long short-term memory networks and interpretable models
• Learning Convolutional Text Representations for Visual Question Answering
• Robust Lottery Tickets for Pre-trained Language Models
• Making Use of NXt to Nothing: The Effect of Class Imbalances on DGA Detection Classifiers
• Character-level HyperNetworks for Hate Speech Detection
• Using Machine Learning and Natural Language Processing to Review and Classify the Medical Literature on Cancer Susceptibility Genes
• Neural Language Model Based Training Data Augmentation for Weakly Supervised Early Rumor Detection
• Weakly-Supervised Scientific Document Classification via Retrieval-Augmented Multi-Stage Training
• NewsEmbed: Modeling News through Pre-trained Document Representations
• MICE: Mining Idioms with Contextual Embeddings
• Small-Text: Active Learning for Text Classification in Python
• KDk: A Defense Mechanism Against Label Inference Attacks in Vertical Federated Learning
• Deep-Energy: Unsupervised Training of Deep Neural Networks
• A Data-driven Neural Network Architecture for Sentiment Analysis
• LMSanitator: Defending Prompt-Tuning Against Task-Agnostic Backdoors
• Influence-guided Data Augmentation for Neural Tensor Completion
• Low Precision Decentralized Distributed Training over IID and non-IID Data
• vONTSS: vMF based semi-supervised neural topic modeling with optimal transport
• An Ensemble Deep Learning Model for Drug Abuse Detection in Sparse Twitter-Sphere
• Data Distillation for Text Classification
• DeepLens: Interactive Out-of-distribution Data Detection in NLP Models
• Federated Learning on Non-iid Data via Local and Global Distillation
• Domain Adaptation with Category Attention Network for Deep Sentiment Analysis
• MixUp Training Leads to Reduced Overfitting and Improved Calibration for the Transformer Architecture
• Towards Practical Few-shot Federated NLP
• Towards Metric DBSCAN: Exact, Approximate, and Streaming Algorithms
• POSNoise: An Effective Countermeasure Against Topic Biases in Authorship Analysis
• The Dark Side of the Language: Pre-trained Transformers in the DarkNet
• Connecting the Dots: What Graph-Based Text Representations Work Best for Text Classification Using Graph Neural Networks?
• VendorLink: An NLP approach for Identifying & Linking Vendor Migrants & Potential Aliases on Darknet Markets
• Abstractive Summarization for Low Resource Data using Domain Transfer and Data Synthesis
• An Attention-Gated Convolutional Neural Network for Sentence Classification
• DeepSeer: Interactive RNN Explanation and Debugging via State Abstraction
• Improving the Robustness of Transformer-based Large Language Models with Dynamic Attention
• On the Impact of Knowledge-based Linguistic Annotations in the Quality of Scientific Embeddings
• Investigating Deep Learning Approaches for Hate Speech Detection in Social Media
• Short Text Classification via Knowledge powered Attention with Similarity Matrix based CNN
• $k$Folden: $k$-Fold Ensemble for Out-Of-Distribution Detection
• SMILE: Evaluation and Domain Adaptation for Social Media Language Understanding
• Do sequence-to-sequence VAEs learn global features of sentences?
• Bayesian Multi-Task Transfer Learning for Soft Prompt Tuning
• DRILL: Dynamic Representations for Imbalanced Lifelong Learning
• Local Boosting for Weakly-Supervised Learning
• Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches
• NLP Verification: Towards a General Methodology for Certifying Robustness
• This Reads Like That: Deep Learning for Interpretable Natural Language Processing
• Weakly Supervised Reasoning by Neuro-Symbolic Approaches
• Adaptive Weight Assignment Scheme For Multi-task Learning
• Preserving Semantics in Textual Adversarial Attacks
• Collaborative Intelligence Orchestration: Inconsistency-Based Fusion of Semi-Supervised Learning and Active Learning
• End-to-End Text Classification via Image-based Embedding using Character-level Networks
• TextNAS: A Neural Architecture Search Space tailored for Text Representation
• On Sampling Strategies for Neural Network-based Collaborative Filtering
• Interpreting Deep Models through the Lens of Data
• What's in a Name? -- Gender Classification of Names with Character Based Machine Learning Models
• Disambiguation of morpho-syntactic features of African American English -- the case of habitual be
• Which Discriminator for Cooperative Text Generation?
• Marich: A Query-efficient Distributionally Equivalent Model Extraction Attack using Public Data
• Universal Rules for Fooling Deep Neural Networks based Text Classification
• 2kenize: Tying Subword Sequences for Chinese Script Conversion
• Augmenting Slot Values and Contexts for Spoken Language Understanding with Pretrained Models
• Uncovering the Connections Between Adversarial Transferability and Knowledge Transferability
• GPT3Mix: Leveraging Large-scale Language Models for Text Augmentation
• A low latency ASR-free end to end spoken language understanding system
• Scalable Acceleration for Classification-Based Derivative-Free Optimization
• Lost In Translation: Generating Adversarial Examples Robust to Round-Trip Translation
• AdvCodeMix: Adversarial Attack on Code-Mixed Data
• Graphmax for Text Generation
• Authorship Attribution in Bangla literature using Character-level CNN
• XC-NAS: A New Cellular Encoding Approach for Neural Architecture Search of Multi-path Convolutional Neural Networks
• Learning Representations for Axis-Aligned Decision Forests through Input Perturbation
• R&R: Metric-guided Adversarial Sentence Generation
• Vec2GC -- A Graph Based Clustering Method for Text Representations
• Transformer-F: A Transformer network with effective methods for learning universal sentence representation
• AMPLIFY:Attention-based Mixup for Performance Improvement and Label Smoothing in Transformer
• Automatic Intent-Slot Induction for Dialogue Systems
• Linguistic Knowledge in Data Augmentation for Natural Language Processing: An Example on Chinese Question Matching
• Pre-train, Interact, Fine-tune: A Novel Interaction Representation for Text Classification
• Text Classification based on Multiple Block Convolutional Highways
• Ranking Entity Based on Both of Word Frequency and Word Sematic Features
• Learning Neural Networks on SVD Boosted Latent Spaces for Semantic Classification
• MyMigrationBot: A Cloud-based Facebook Social Chatbot for Migrant Populations
• The emergent algebraic structure of RNNs and embeddings in NLP