
Natural Language Processing and Chinese Computing
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
The 62 full papers, 21 poster papers, and 27 workshop papers presented were carefully reviewed and selected from 327 submissions. They are organized in the following areas: Fundamentals of NLP; Machine Translation and Multilinguality; Machine Learning for NLP; Information Extraction and Knowledge Graph; Summarization and Generation; Question Answering; Dialogue Systems; Social Media and Sentiment Analysis; NLP Applications and Text Mining; and Multimodality and Explainability.
More details
Other editions
Additional editions

Content
- Intro
- Preface
- Organization
- Contents - Part II
- Contents - Part I
- Question Answering (Poster)
- Faster and Better Grammar-Based Text-to-SQL Parsing via Clause-Level Parallel Decoding and Alignment Loss
- 1 Introduction
- 2 Related Works
- 3 Our Proposed Model
- 3.1 Grammar-Based Text-to-SQL Parsing
- 3.2 Clause-Level Parallel Decoding
- 3.3 Clause-Level Alignment Loss
- 4 Experiments
- 4.1 Experimental Setup
- 4.2 Results
- 4.3 Analysis
- 5 Conclusions
- References
- Two-Stage Query Graph Selection for Knowledge Base Question Answering
- 1 Introduction
- 2 Our Approach
- 2.1 Query Graph Generation
- 2.2 Two-Stage Query Graph Selection
- 3 Experiments
- 3.1 Experimental Setup
- 3.2 Main Results
- 3.3 Discussion and Analysis
- 4 Related Work
- 5 Conclusions
- References
- Plug-and-Play Module for Commonsense Reasoning in Machine Reading Comprehension
- 1 Introduction
- 2 Methodology
- 2.1 Task Formulation
- 2.2 Proposed Module: PIECER
- 2.3 Plugging PIECER into MRC Models
- 3 Experiments
- 3.1 Datasets
- 3.2 Base Models
- 3.3 Experimental Settings
- 3.4 Main Results
- 3.5 Analysis and Discussions
- 4 Related Work
- 5 Conclusion
- References
- Social Media and Sentiment Analysis (Poster)
- FuDFEND: Fuzzy-Domain for Multi-domain Fake News Detection
- 1 Introduction
- 2 Related Work
- 2.1 Fake News Detection Methods
- 2.2 Multi-domain Rumor Task
- 3 FuDFEND: Fuzzy-Domain Fake News Detection Model
- 3.1 Membership Function
- 3.2 Feature Extraction
- 3.3 Domain Gate
- 3.4 Fake News Prediction and Loss Function
- 4 Experiment
- 4.1 Dataset
- 4.2 Experiment Setting
- 4.3 Train Membership Function and FuDFEND
- 4.4 Experiment on Weibo21
- 4.5 Experiment on Thu Dataset
- 5 Conclusion
- 6 Future Work
- References
- NLP Applications and Text Mining (Poster)
- Continuous Prompt Enhanced Biomedical Entity Normalization
- 1 Introduction
- 2 Related Work
- 2.1 Biomedical Entity Normalization
- 2.2 Prompt Learning and Contrastive Loss
- 3 Our Method
- 3.1 Prompt Enhanced Scoring Mechanism
- 3.2 Contrastive Loss Enhanced Training Mechanism
- 4 Experiments and Analysis
- 4.1 Dataset and Evaluation
- 4.2 Data Preprocessing
- 4.3 Experiment Setting
- 4.4 Overall Performance
- 4.5 Ablation Study
- 5 Conclusion
- References
- Bidirectional Multi-channel Semantic Interaction Model of Labels and Texts for Text Classification
- 1 Introduction
- 2 Model
- 2.1 Preliminaries
- 2.2 Bidirectional Multi-channel Semantic Interaction Model
- 3 Experiments
- 3.1 Experimental Settings
- 3.2 Results and Analysis
- 3.3 Ablation Test
- 4 Conclusions
- References
- Exploiting Dynamic and Fine-grained Semantic Scope for Extreme Multi-label Text Classification
- 1 Introduction
- 2 Related Work
- 3 Methodology
- 3.1 Notation
- 3.2 TReaderXML
- 4 Experiments
- 4.1 Datasets and Preprocessing
- 4.2 Baselines
- 4.3 Evaluation Metrics
- 4.4 Ablation Study
- 4.5 Performance on Tail Labels
- 5 Conclusions
- References
- MGEDR: A Molecular Graph Encoder for Drug Recommendation
- 1 Introduction
- 2 Related Works
- 2.1 Drug Recommendation
- 2.2 Molecular Graph Representation
- 3 Problem Formulation
- 4 The MGEDR Model
- 4.1 Patient Encoder
- 4.2 Medicine Encoder
- 4.3 Functional Groups Encoder
- 4.4 Medicine Representation
- 4.5 Optimization
- 5 Experiments
- 5.1 Dataset and Metrics
- 5.2 Results
- 5.3 Ablations
- 6 Conclusion
- References
- Student Workshop (Poster)
- Semi-supervised Protein-Protein Interactions Extraction Method Based on Label Propagation and Sentence Embedding
- 1 Introduction
- 2 Related Work
- 3 Methods
- 3.1 Problem Formulation
- 3.2 Overall Workflow
- 3.3 Label Propagation
- 3.4 Sentence Embedding
- 3.5 CNN Classifier
- 4 Results
- 4.1 Datasets and Preprocessing
- 4.2 Experimental Results
- 4.3 Hyperparameter Analysis
- 5 Conclusion
- References
- Construction and Application of a Large-Scale Chinese Abstractness Lexicon Based on Word Similarity
- 1 Introduction
- 2 Data and Method
- 2.1 Data
- 2.2 Method
- 3 Experiment
- 4 Construction and Evaluation
- 5 Application
- 5.1 Cross-Language Comparison
- 5.2 Chinese Text Readability Auto-evaluation
- 6 Conclusion
- References
- Stepwise Masking: A Masking Strategy Based on Stepwise Regression for Pre-training
- 1 Introduction
- 2 Methodology
- 2.1 Three-Stage Framework
- 2.2 Stepwise Masking
- 3 Experiments
- 3.1 Datasets
- 3.2 Experimental Settings
- 3.3 Main Results
- 3.4 Effectiveness of Stepwise Masking
- 3.5 Effect of Dynamic in Stepwise Masking
- 3.6 Case Study
- 4 Conclusion and Future Work
- References
- Evaluation Workshop (Poster)
- Context Enhanced and Data Augmented W2NER System for Named Entity Recognition
- 1 Introduction
- 2 Related Work
- 3 The Proposed Approach
- 3.1 Task Definition
- 3.2 Model Structure
- 3.3 Data Augmentation
- 3.4 Result Ensemble
- 4 Experiments
- 4.1 Dataset and Metric
- 4.2 Experiment Settings
- 4.3 Baselines
- 4.4 Results and Analysis
- 5 Conclusion
- References
- Multi-task Hierarchical Cross-Attention Network for Multi-label Text Classification
- 1 Introduction
- 2 Related Work
- 2.1 Hierarchical Multi-label Text Classification
- 2.2 Representation of Scientific Literature
- 3 Methodology
- 3.1 Representation Layer
- 3.2 Hierarchical Cross-Attention Recursive Layer
- 3.3 Hierarchical Prediction Layer
- 3.4 Rebalanced Loss Function
- 4 Experiment
- 4.1 Dataset and Evaluation
- 4.2 Experimental Settings
- 4.3 Results and Discussions
- 4.4 Module Analysis
- 5 Conclusion
- References
- An Interactive Fusion Model for Hierarchical Multi-label Text Classification
- 1 Introduction
- 2 Related Work
- 3 Task Definition
- 4 Method
- 4.1 Shared Encoder Module
- 4.2 Task-Specific Module
- 4.3 Training and Inference
- 5 Experiment
- 6 Conclusion
- References
- Scene-Aware Prompt for Multi-modal Dialogue Understanding and Generation
- 1 Introduction
- 2 Task Introduction
- 2.1 Problem Definition
- 2.2 Evaluation Metric
- 2.3 Dateset
- 3 Main Methods
- 3.1 Multi-tasking Multi-modal Dialogue Understanding
- 3.2 Scene-Aware Prompt Multi-modal Dialogue Generation
- 3.3 Training and Inference
- 4 Experiments
- 4.1 Experimental Setup
- 4.2 Main Results
- 4.3 Ablation Study
- 4.4 Online Results
- 5 Conclusion
- References
- BIT-WOW at NLPCC-2022 Task5 Track1: Hierarchical Multi-label Classification via Label-Aware Graph Convolutional Network
- 1 Introduction
- 2 Approach
- 2.1 Context-Aware Label Embedding
- 2.2 Graph-Based Hierarchical Label Modeling
- 2.3 Curriculum Learning Strategy
- 2.4 Ensemble Learning and Post Editing
- 3 Experiments
- 3.1 Dataset and Experiment Settings
- 3.2 Main Results
- 3.3 Analysis
- 4 Related Work
- 5 Conclusion
- References
- CDAIL-BIAS MEASURER: A Model Ensemble Approach for Dialogue Social Bias Measurement
- 1 Introduction
- 2 Related Work
- 2.1 Shared Tasks
- 2.2 Solution Models
- 3 Dataset
- 4 Method
- 4.1 Models Selection
- 4.2 Fine-Tuning Strategies
- 4.3 Ensembling Strategy
- 5 Result
- 5.1 Preliminary Screening
- 5.2 Model Ensemble
- 5.3 Ensemble Size Effect
- 5.4 Discussion
- 6 Conclusion
- References
- A Pre-trained Language Model for Medical Question Answering Based on Domain Adaption
- 1 Introduction
- 2 Related Work
- 2.1 Encoder-Based
- 2.2 Decoder-Based
- 2.3 Encoder-Decoder-Based
- 3 Description of the Competition
- 3.1 Evaluation Metrics
- 3.2 Datasets
- 4 Solution
- 4.1 Model Introduction
- 4.2 Strategy
- 4.3 Model Optimization
- 4.4 Model Evaluation
- 5 Conclusion
- References
- Enhancing Entity Linking with Contextualized Entity Embeddings
- 1 Introduction
- 2 Related Work
- 3 Methodology
- 3.1 Dual Encoder
- 3.2 LUKE-Based Cross-Encoder
- 4 Experiments
- 4.1 Data
- 4.2 Candidate Retrieval
- 4.3 Candidate Reranking
- 5 Conclusion
- References
- A Fine-Grained Social Bias Measurement Framework for Open-Domain Dialogue Systems
- 1 Introduction
- 2 Related Work
- 2.1 Fine Grained Dialogue Social Bias Measurement
- 2.2 Application of Contrastive Learning in NLP Tasks
- 2.3 Application of Prompt Learning in NLP Tasks
- 3 Fine-Grain Dialogue Social Bias Measurement Framework
- 3.1 General Representation Module
- 3.2 Two-Stage Prompt Learning Module
- 3.3 Contrastive Learning Module
- 4 Experiment
- 4.1 Dataset
- 4.2 Experimental Setup
- 4.3 Results and Analysis
- 5 Conclusion
- References
- Dialogue Topic Extraction as Sentence Sequence Labeling
- 1 Introduction
- 2 Related Work
- 2.1 Dialogue Topic Information
- 2.2 Sequence Labeling
- 3 Methodology
- 3.1 Task Definition
- 3.2 Topic Extraction Model
- 3.3 Ensemble Model
- 4 Experiments
- 4.1 Dataset
- 4.2 Results and Analysis
- 5 Conclusion
- References
- Knowledge Enhanced Pre-trained Language Model for Product Summarization
- 1 Introduction
- 2 Related Work
- 2.1 Encoder-Decoder Transformer
- 2.2 Decoder-Only Transformer
- 3 Description of the Competition
- 4 Dataset Introduction
- 4.1 Textual Data
- 4.2 Image Data
- 5 Model Solution
- 5.1 Model Introduction
- 5.2 Model Training
- 6 Model Evaluation
- 7 Conclusion
- References
- Augmented Topic-Specific Summarization for Domain Dialogue Text
- 1 Introduction
- 2 Related Work
- 3 Dataset
- 3.1 Data Preprocessing
- 3.2 Constrained Search of Training Data
- 4 System Overview
- 4.1 Sub-summary Generation Model
- 4.2 Ensemble Topic Detection
- 4.3 Data Augmentation
- 5 Experiments
- 6 Analysis
- 6.1 The Effect of Different Data Augmentation Methods
- 6.2 The Effect of Classification
- 7 Conclusion
- References
- DAMO-NLP at NLPCC-2022 Task 2: Knowledge Enhanced Robust NER for Speech Entity Linking
- 1 Introduction
- 2 Related Work
- 3 Methods
- 3.1 Candidate Entity Retriever
- 3.2 Knowledge Enhanced NER
- 3.3 Linking with Filtering
- 3.4 Ensemble Methods
- 4 Experiments
- 4.1 Data Analysis and Experimental Settings
- 4.2 Main Results
- 4.3 Ablation Study
- 5 Conclusions
- References
- Overview of the NLPCC2022 Shared Task on Speech Entity Linking
- 1 Introduction
- 2 Related Work
- 3 Task Description
- 3.1 Track 1: Entity Recognition and Disambiguation
- 3.2 Track 2: Entity Disambiguation-Only
- 4 Dataset Description
- 5 Result
- 5.1 Evaluation Metrics
- 5.2 Participants and Results
- 5.3 Analysis
- 6 Conclusion
- References
- Overview of the NLPCC 2022 Shared Task on Multimodal Product Summarization
- 1 Introduction
- 2 Task Definition
- 3 Dataset
- 4 Evaluation
- 5 Participants
- 6 Results
- 6.1 Automatic Evaluation Results
- 6.2 Human Evaluation Results
- 7 Conclusion
- A An Example from the Dataset
- References
- A Multi-task Learning Model for Fine-Grain Dialogue Social Bias Measurement
- 1 Introduction
- 2 Related Work
- 3 Proposed System
- 3.1 Problem Definition
- 3.2 Text Representation
- 3.3 Auxiliary Tasks
- 3.4 Main Task
- 3.5 Model Fusion
- 3.6 Adversarial Training
- 3.7 Training Loss
- 4 Experiments and Details
- 4.1 Dataset
- 4.2 Implement Details
- 4.3 Results and Analysis
- 5 Conclusion
- References
- Overview of NLPCC2022 Shared Task 5 Track 1: Multi-label Classification for Scientific Literature
- 1 Introduction
- 2 Related Work
- 3 Data
- 4 Methods from Participants
- 5 Results
- 6 Conclusion
- References
- Overview of the NLPCC 2022 Shared Task: Multi-modal Dialogue Understanding and Generation
- 1 Introduction
- 2 Task Description
- 2.1 Dialogue Scene Identification
- 2.2 Dialogue Session Identification
- 2.3 Dialogue Response Generation
- 3 Dataset Description
- 4 Results
- 4.1 Evaluation Metrics
- 4.2 Evaluation Results
- 5 Conclusion
- References
- Overview of NLPCC2022 Shared Task 5 Track 2: Named Entity Recognition
- 1 Introduction
- 2 Task Definition
- 3 Dataset Preparation
- 4 Evaluation Metric
- 5 Approaches and Results
- 6 Conclusion
- References
- Overview of NLPCC 2022 Shared Task 7: Fine-Grained Dialogue Social Bias Measurement
- 1 Introduction
- 2 Task Description
- 2.1 Task Formulation
- 2.2 Data Collection
- 2.3 Annotation Schema
- 2.4 Human Annotation
- 3 Evaluation Results
- 3.1 Evaluation Metrics
- 3.2 Submission Results
- 4 Representative Systems
- 5 Conclusion
- References
- Overview of the NLPCC 2022 Shared Task: Dialogue Text Analysis (DTA)
- 1 Introduction
- 2 Task Description
- 2.1 Dialogue Topic Extraction (DTE)
- 2.2 Dialogue Summary Generation (DSG)
- 3 Data Construction
- 4 Evaluation Metrics
- 5 Evaluation Results
- 6 Conclusion
- References
- Author Index
System requirements
File format: PDF
Copy protection: Watermark-DRM (Digital Rights Management)
System requirements:
- Computer (Windows; MacOS X; Linux): Use the free software Adobe Reader, Adobe Digital Editions, or any other PDF viewer of your choice (see eBook Help).
- Tablet/Smartphone (Android; iOS): Install the free app Adobe Digital Editions or another reading app for eBooks, e.g., PocketBook (see eBook Help).
- E-reader: Bookeen, Kobo, Pocketbook, Sony, Tolino and many more (only limited: Kindle).
The file format PDF always displays a book page identically on any hardware. This makes PDF suitable for complex layouts such as those used in textbooks and reference books (images, tables, columns, footnotes). Unfortunately, on the small screens of e-readers or smartphones, PDFs are rather annoying, requiring too much scrolling.
This eBook uses Watermark-DRM, a „soft” copy protection. This means that there are no technical restrictions to prevent illegal distribution. However, there is a personalised watermark embedded in the eBook that can be used to identify the purchaser of the eBook in the event of misuse and to provide evidence for legal purposes.
For more information, see our eBook Help page.