Convolutional Neural Network Architectures for Matching Natural Language Sentences

Abstract

Semantic matching is of central importance to many natural language tasks \cite{bordes2014semantic,RetrievalQA}. A successful matching algorithm needs to adequately model the internal structures of language objects and the interaction between them. As a step toward this goal, we propose convolutional neural network models for matching two sentences, by adapting the convolutional strategy in vision and speech. The proposed models not only nicely represent the hierarchical structures of sentences with their layer-by-layer composition and pooling, but also capture the rich matching patterns at different levels. Our models are rather generic, requiring no prior knowledge on language, and can hence be applied to matching tasks of different nature and in different languages. The empirical study on a variety of matching tasks demonstrates the efficacy of the proposed model on a variety of matching tasks and its superiority to competitor models.

References in corpus (2)

• Improving neural networks by preventing co-adaptation of feature detectors
• Natural Language Processing (almost) from Scratch

Cited by in corpus (55)

• A Deep Relevance Matching Model for Ad-hoc Retrieval
• End-to-End Neural Ad-hoc Ranking with Kernel Pooling
• Representation learning for very short texts using weighted word embedding aggregation
• Question Answering and Question Generation as Dual Tasks
• Adversarial Feature Matching for Text Generation
• Classifying Relations by Ranking with Convolutional Neural Networks
• Extraction of Salient Sentences from Labelled Documents
• Deconvolutional Paragraph Representation Learning
• Discriminative Neural Sentence Modeling by Tree-Based Convolution
• Natural Language Understanding with the Quora Question Pairs Dataset
• Encoding Source Language with Convolutional Neural Network for Machine Translation
• Self-Adaptive Hierarchical Sentence Model
• Learning text representation using recurrent convolutional neural network with highway layers
• Neural Information Retrieval: A Literature Review
• Multiresolution Graph Attention Networks for Relevance Matching
• MatchZoo: A Toolkit for Deep Text Matching
• Deep Multi-Task Learning with Shared Memory
• Syntax-based Deep Matching of Short Texts
• CNM: An Interpretable Complex-valued Network for Matching
• Answer Sequence Learning with Neural Networks for Answer Selection in Community Question Answering
• Incorporating Query Term Independence Assumption for Efficient Retrieval and Ranking using Deep Neural Networks
• Local Translation Prediction with Global Sentence Representation
• A Deep Look into Neural Ranking Models for Information Retrieval
• Fine-grained Event Categorization with Heterogeneous Graph Convolutional Networks
• Learning to Match Using Local and Distributed Representations of Text for Web Search
• Semantic Product Search
• Text-to-Viz: Automatic Generation of Infographics from Proportion-Related Natural Language Statements
• $gen$CNN: A Convolutional Architecture for Word Sequence Prediction
• A Deep Investigation of Deep IR Models
• Automated Steel Bar Counting and Center Localization with Convolutional Neural Networks
• DAL: Dual Adversarial Learning for Dialogue Generation
• Context-Dependent Translation Selection Using Convolutional Neural Network
• Cross Temporal Recurrent Networks for Ranking Question Answer Pairs
• Neural IR Meets Graph Embedding: A Ranking Model for Product Search
• Learning a Matching Model with Co-teaching for Multi-turn Response Selection in Retrieval-based Dialogue Systems
• Toward a Deep Neural Approach for Knowledge-Based IR
• A Context-Aware User-Item Representation Learning for Item Recommendation
• Deep Binaries: Encoding Semantic-Rich Cues for Efficient Textual-Visual Cross Retrieval
• Co-PACRR: A Context-Aware Neural IR Model for Ad-hoc Retrieval
• Promotion of Answer Value Measurement with Domain Effects in Community Question Answering Systems
• Building Data-driven Models with Microstructural Images: Generalization and Interpretability
• A Sequential Matching Framework for Multi-turn Response Selection in Retrieval-based Chatbots
• Learning Semantically Coherent and Reusable Kernels in Convolution Neural Nets for Sentence Classification
• A Document-grounded Matching Network for Response Selection in Retrieval-based Chatbots
• Neural Network Based Next-Song Recommendation
• Lattice CNNs for Matching Based Chinese Question Answering
• Context Attentive Document Ranking and Query Suggestion
• Modelling Domain Relationships for Transfer Learning on Retrieval-based Question Answering Systems in E-commerce
• Impact of Training Dataset Size on Neural Answer Selection Models
• Luandri: a Clean Lua Interface to the Indri Search Engine
• TraceWalk: Semantic-based Process Graph Embedding for Consistency Checking
• Learning Word Embeddings from Intrinsic and Extrinsic Views
• Character-level Deep Conflation for Business Data Analytics
• Learning Fast Matching Models from Weak Annotations
• Automatic Question-Answering Using A Deep Similarity Neural Network