
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.
This book constitutes the refereed proceedings of the 4th CCF Conference, NLPCC 2015, held in Nanchang, China, in October 2015.
The 35 revised full papers presented together with 22 short papers were carefully reviewed and selected from 238 submissions. The papers are organized in topical sections on fundamentals on language computing; applications on language computing; NLP for search technology and ads; web mining; knowledge acquisition and information extraction.
More details
Other editions
Additional editions

Content
- Intro
- Preface
- Organization
- Contents
- Fundamentals on Language Computing
- A Maximum Entropy Approach to Discourse Coherence Modeling
- 1 Introduction
- 2 Related Work
- 3 Maximum Entropy Based Discourse Coherence Model
- 3.1 Discourse Model
- 3.2 Maximum Entropy Based Discourse Coherence Model
- 3.3 Model Training
- 4 Experiment
- 4.1 Dataset
- 4.2 Model Comparison
- 5 Conclusion
- References
- Transition-Based Dependency Parsing with Long Distance Collocations
- 1 Introduction
- 2 Transition-Based Dependency Parsing
- 3 Incorporating Long Distance Collocations
- 3.1 Collocations Extraction
- 3.2 Features with Collocation Information
- 3.3 Speedup
- 4 Enhanced Actions with Fine-Grained ``Shift''
- 5 Experiments
- 5.1 Datasets
- 5.2 Parsing Accuracy
- 5.3 Parsing Time
- 5.4 Result Analysis with Length Factors
- 5.5 Result Analysis with Different POS Tags
- 6 Related Works
- 7 Conclusion
- References
- Recurrent Neural Networks with External Memory for Spoken Language Understanding
- 1 Introduction
- 2 Background
- 2.1 Language Understanding
- 2.2 Simple Recurrent Neural Networks
- 2.3 Recurrent Neural Networks Using Gating Functions
- 3 The RNN-EM Architecture
- 3.1 Model Input and Output
- 3.2 External Memory Read
- 3.3 External Memory Update
- 3.4 Dataset
- 3.5 Comparison with the Past Results
- 3.6 Analysis on Convergence and Averaged Performances
- 3.7 Analysis on Memory Size
- 4 Related Work
- 5 Conclusion and Discussion
- Improving Chinese Dependency Parsing with Lexical Semantic Features
- 1 Introduction
- 2 Framework
- 2.1 Extracting Semantic Categories
- 2.2 Semantic Feature Templates
- 2.3 Semantic Similarity
- 3 Experiment
- 3.1 Data set
- 3.2 Experimental Results
- 4 Related Work
- Conclusion
- References
- Machine Translation and Multi-Lingual Information Access
- Entity Translation with Collective Inference in Knowledge Graph
- 1 Introduction
- 2 Related Work
- 3 Our Method
- 3.1 Multi-Source Candidates Generation
- 3.2 Collective Inference with Neighbors
- 3.3 Translation Graph
- 3.4 Features
- 3.5 Training
- 4 Experiments
- 4.1 Dataset
- 4.2 Baselines
- 4.3 Translation Result
- 4.4 Training Phase Gain
- 4.5 Feature Gain
- 4.6 Source Gain
- 5 Conclusion
- Stochastic Language Generation Using Situated PCFGs
- 1 Introduction
- 2 Related Work
- 3 Problem Formulation
- 3.1 The Grammar
- 3.2 Grammar Induction
- 3.3 Decoding
- 4 Empirical Evaluation
- 4.1 Data Set
- 4.2 Evaluation Metric
- 4.3 Results
- 5 Conclusions
- References
- Machine Learning for NLP
- Clustering Sentiment Phrases in Product Reviews by Constrained Co-clustering
- 1 Introduction
- 2 Methodology
- 2.1 Information-Theoretic Co-clustering
- 2.2 Constrained Co-clustering
- 2.3 Constraints
- 2.4 Sentiment Phrase Extraction
- 3 Experiment Results
- 3.1 Data Preparation
- 3.2 Evaluation Metrics
- 3.3 Evaluation Results
- 4 Related Works
- 5 Conclusion
- References
- A Cross-Domain Sentiment Classification Method Based on Extraction of Key Sentiment Sentence
- 1 Introduction
- 2 Problem Setting and Related Concepts
- 2.1 Problem Setting
- 2.2 Related Concepts
- 3 Cross-Domian Sentiment Classification Method Based on Extraction of Key Sentiment Sentence
- 3.1 Key Sentiment Sentence Extraction Algorithm
- Sentiment Purity.
- Keyword Property.
- Position Property.
- 3.2 Multi-view Ensemble
- 4 Experiments
- 4.1 Dataset Description
- 4.2 Baselines Setting
- 4.3 Overall Comparision Results
- 4.4 Parameter Sensitivity Analysis
- 5 Conclusion
- References
- Convolutional Neural Networks for Correcting English Article Errors
- 1 Introduction
- 2 Related Work
- 3 Model
- 3.1 Model Architecture
- 3.2 Preprocessing Module
- 3.3 CNN Module
- 3.4 Postprocessing Module
- 4 Dataset and Evaluation Metrics
- 5 Experiments
- 5.1 Pre-trained Word Embeddings
- 5.2 Parameter Settings
- 5.3 Experiment Results
- 6 Conclusion and Feature Work
- References
- NLP for Social Media
- Automatic Detection of Rumor on Social Network
- 1 Introduction
- 2 Related Works
- 3 Proposed Method
- 3.1 Rumor Detection Flow
- 3.2 Content-Based Implicit Features
- 3.3 User-Based Implicit Features
- 3.4 Feature Fusion
- 4 Experiments
- 4.1 Dataset
- 4.2 Evaluation
- 5 Conclusion
- References
- Multimodal Learning Based Approaches for Link Prediction in Social Networks
- 1 Introduction
- 2 Background
- 2.1 Restricted Boltzmann Machine
- 2.2 Deep Belief Network
- 3 Link Prediction Problem
- 4 Methodology
- 4.1 Link Network Structure Features
- 4.2 User Comment Features
- 4.3 Discriminative Deep Belief Networks
- 4.4 Multimodal Deep Belief Networks
- 5 Experiments and Analysis
- 5.1 Experiment Setup
- 5.2 Experiment Results and Analysis
- 6 Conclusion
- References
- Sentiment Analysis Based on User Tags for Traditional Chinese Medicine in Weibo
- 1 Introduction
- 2 Data Collecting and Labelling
- 2.1 Corpus Collection Based on User Tags
- 2.2 Two Dictionary Resources
- 2.3 Pre-processing of Data
- 2.4 Filtering Chinese Medicine Tweets
- 2.5 Labelling the Data
- 3 Methodology
- 3.1 Feature Selection
- 3.2 Machine Learning Method
- 3.3 Adjusting Sentiment Classification Results
- 4 Experiments and Results
- 4.1 The Performance Measure
- 4.2 Feature Selection Results
- 4.3 Classification Results
- 4.4 Adjusted Classification Results
- 4.5 Prediction
- 5 Conclusion and Future Work
- References
- Predicting User Mention Behavior in Social Networks
- 1 Introduction
- 2 Related Work
- 3 Problem Statement
- 4 User Mention Prediction Model
- 4.1 Feature Extraction
- 4.2 Mention Behavior Prediction
- 5 Experiments and Analysis
- 5.1 Data Collection
- 5.2 Comparison Methods and Evaluation Metrics
- 5.3 Results and Discussion
- 6 Conclusions and Future Work
- References
- Convolutional Neural Networks for Multimedia Sentiment Analysis
- 1 Introduction
- 2 Related Work
- 2.1 Textual Sentiment Analysis
- 2.2 Visual Sentiment Analysis
- 2.3 Multimedia Sentiment Analysis
- 3 Textual and Visual Sentiment Analysis with CNN
- 3.1 Textual Sentiment Analysis with CNN
- 3.2 Visual Sentiment Analysis with CNN
- 3.3 Multimedia Sentiment Analysis with CNN
- 3.4 Classification
- 4 Experimental Setup and Results
- 4.1 Datasets
- 4.2 Pre-trained Word Vectors
- 4.3 CNN Training
- 4.4 Results
- 5 Conclusions
- References
- Applications on Language Computing
- An Adaptive Approach to Extract Characters from Digital Ink Text in Chinese Based on Extracted Errors
- 1 Introduction
- 2 Related Works
- 3 Classification of the Extraction Errors
- 3.1 Deficient Extraction
- 3.2 Beyond Extraction
- 3.3 False Extraction
- 3.4 Classifying Rules
- 4 Method for Deficient Extraction Errors
- 4.1 Eigenvectors Extraction
- 4.2 Hierarchical Agglomeration Clustering Algorithm
- 4.3 K-means Clusterin g Algorithm
- 5 Method for Beyond Extraction
- 6 Performance Test
- 7 Conclusions
- Reference
- Context-Dependent Metaphor Interpretation Based on Semantic Relatedness
- 1 Introduction
- 2 Related Work
- 2.1 Metaphor Interpretation
- 2.2 Semantic Relatedness
- 3 Our Method
- 3.1 Theoretical Basis
- 3.2 Metaphor Interpretation Based on Semantic Relatedness
- 4 Experiment and Evaluation
- 5 Conclusions
- References
- Context Vector Model for Document Representation: A Computational Study
- 1 Introduction
- 2 Preliminaries
- 2.1 The BOW Model
- 2.2 Context Vector Model
- 2.3 Generating the Word Vectors
- 3 Experimental Setup
- 3.1 Dataset
- 3.2 Methods to Be Compared
- 3.3 Similarity and Distance Measure
- 3.4 Evaluation Metrics
- Overlap Ratio.
- Standard Deviation.
- F1-Score.
- Normalized Mutual Information
- 3.5 Clustering Method
- 4 Experimental Results
- 4.1 Discrimination of Document Vectors
- 4.2 Performance on Document Clustering
- 4.3 Discussion about Word Representation Methods
- 5 Conclusions
- References
- NLP for Search Technology and Ads
- Refine Search Results Based on Desktop Context
- 1 Motivations
- 2 Related Works
- 3 Desktop Context Model
- 3.1 Analyzing Document Information
- 3.2 Analyzing Work Scenario
- 4 Experimental Evaluation
- 5 Conclusion
- References
- Incorporating Semantic Knowledge with MRF Term Dependency Model in Medical Document Retrieval
- 1 Introduction
- 2 Related Work
- 3 Methodology
- 3.1 Concept Detection
- 3.2 Concept Relation Extraction
- 3.3 Feature Function
- 4 Experiments
- 4.1 Experiment 1: Feature Function for Single Medical Concept
- 4.2 Experiment 2: Results of Different Models
- 5 Conclusion
- References
- A Full-Text Retrieval Algorithm for Encrypted Data in Cloud Storage Applications
- 1 Introduction
- 2 Related Work
- 2.1 Keyword Searchable Encryption
- 2.2 Rich Functional Encrypted Data Search
- 2.3 Ranked Search Over Encrypted Data
- 3 Full-Text Retrieval Algorithm Over Encrypted Data
- 3.1 Full-Text Retrieval Model and Its Security Problem
- 3.2 System Model
- 3.3 Privacy-Preserved Full-Text Retrieval Index Based on B+ Tree
- 3.4 Document Pre-processing
- 3.5 Full-Text Retrieval Algorithm Over the Encrypted Data
- 4 Performance and Security Analysis
- 4.1 Query Precision of Our Scheme
- 4.2 Query Efficiency of Our Scheme
- 4.3 Security Analysis of Our Scheme
- 5 Experiments
- 5.1 Experimental Setup
- 5.2 Storage Overhead of Our Scheme
- 5.3 Query Efficiency of Our Scheme
- 6 Conclusion
- References
- How Different Features Contribute to the Session Search?
- 1 Introduction
- 2 Related Work
- 3 Background
- 4 A Classification of Interaction Features
- 4.1 Current Query Features
- 4.2 Query Change Features
- 4.3 Whole Session Features
- 4.4 Collective Intelligence Features
- 5 Empirical Comparison of Different Features
- 5.1 Experimental Setup
- 5.2 Results and Analysis
- 6 Conclusions and Future work
- References
- Web Mining
- Beyond Your Interests: Exploring the Information Behind User Tags
- 1 Introduction
- 2 User Tag Classification in Weibo
- 2.1 User Tag Taxonomy
- 2.2 Feature Extraction for Classification
- Statistical Features.
- Content Features.
- Search Engine Based Syntax Features.
- 2.3 Dataset
- 2.4 Experiments and Results
- 3 User Profiling with Different Generations
- 3.1 User Generation Distribution
- 3.2 Interests Tag Clustering for User Profiling
- 3.3 Result and Analysis
- 4 Area Profiling with Mass Appeal
- 4.1 Area Characteristic Tags Extraction
- 4.2 Evaluation and Analysis
- 5 Conclusion and Future Work
- References
- Nonparametric Symmetric Correspondence Topic Models for Multilingual Text Analysis
- 1 Introduction
- 2 Multilingual Topic Models
- 2.1 Correspondence LDA(CorrLDA)
- 2.2 Symmetric Correspondence LDA(SymCorrLDA)
- 3 Nonparametric Symmetric Correspondence LDA(NPSymCorrLDA)
- 3.1 Dirichlet Process(DP)
- 3.2 NPSymCorrLDA Based on HDP
- 4 Experiments
- 4.1 Settings
- 4.2 Pivot Assignments
- 4.3 Held-Out Log-Likelihood
- 4.4 Finding Counterpart Articles
- 5 Conclusions
- References
- Knowledge Acquisition and Information Extraction
- Mining RDF from Tables in Chinese Encyclopedias
- 1 Introduction
- 2 Approach Overview
- 3 Chinese Knowledge Base from Scratch
- 3.1 Taxonomy Mining
- 3.2 Class Attribute Mining
- 4 Table Classification and Understanding
- 4.1 Column Scoring
- 4.2 Table Classification
- 4.3 RDF Extraction
- 5 Experiments
- 5.1 Column Scoring Evaluation
- 5.2 Table Classification Evaluation
- 5.3 RDF Extraction Evaluation
- 6 Related Work
- 6.1 Knowledge Base Construction
- 6.2 Web Table Understanding
- 7 Conclusion
- References
- Taxonomy Induction from Chinese Encyclopedias by Combinatorial Optimization
- 1 Introduction
- 2 Related Work
- 3 The Proposed Approach
- 3.1 Problem Formulation
- 3.2 Subclass-of Relation Induction
- 3.3 Instance-of Relation Induction
- 4 Experimental Evaluation
- 4.1 Experiment Settings
- 4.2 Performance Evaluation
- 4.3 Parameter Setting
- 5 Conclusion and Future work
- References
- Recognition of Person Relation Indicated by Predicates
- 1 Introduction
- 2 Related Work
- 3 The Model
- 3.1 Transforming Feature Indices into Vectors
- 3.2 Frame Convolution
- 3.3 Dynamic K-max Pooling
- 3.4 One-Dimensional Convolution vs Frame Convolution
- 3.5 Training
- 4 Experiments
- 4.1 Data Set
- 4.2 Experiment-1
- 4.3 Experiment-2
- 4.4 Experiment-3
- 5 Conclusion
- References
- Target Detection and Knowledge Learning for Domain Restricted Question Answering
- 1 Introduction
- 2 Problem Definition
- 2.1 Problem Definition
- 3 Framework
- 3.1 Target-Word Detection
- 3.2 Domain Knowledge Learning
- 3.3 Retrieval Model
- 4 Experiments
- 4.1 Data Preparation
- 4.2 Evaluation Measures
- 4.3 Experiments
- 4.4 Results and Analysis
- 5 Related Work
- 6 Conclusion
- Short Papers
- An Improved Algorithm of Logical Structure Reconstruction for Re-flowable Document Understanding
- 1 Introduction
- 2 The Emergence of Logical Structure Reconstruction Method Based on Directed Graph
- 2.1 Logical Tag Structure Extraction from a Template Document
- 2.2 Construction Method of Directed Graph of a Document
- 2.3 Reconstruction Method of the Logical Structure of a Document
- 3 Experiment and Related Analysis
- 4 Conclusions
- References
- Mongolian Inflection Suffix Processing in NLP: A Case Study
- 1 Introduction
- 2 Inflection Suffix
- Inflection Suffix in Mongolian Syntactic Analysis
- Inflection Suffix in Mongolian Information Retrieval
- Inflection Suffix in Mongolian Translation
- Conclusion
- Reference
- Resolving Coordinate Structures for Chinese Constituent Parsing
- 1 Introduction
- 2 Related Work
- 3 Coordinate Structure in Chinese
- 4 Learning to Resolve Coordinate Structures
- 4.1 Grammar of Chinese Coordinate Structures
- 4.2 New Features
- 5 Experiments
- 5.1 Extraction of Coordinate Structures
- 5.2 Effects on the Parsing Process
- 5.3 Resolving Coordinate Structures
- 6 Conclusion and Future Work
- References
- P-Trie Tree: A Novel Tree Structure for Storing Polysemantic Data
- 1 Introduction
- 1.1 Favorite Words and High Frequency Words
- 1.2 Kuchiguse
- 1.3 NLP and Trie Tree
- 1.4 Polysemy
- 2 Related Works
- 3 P-Trie Tree
- 3.1 Definition
- 3.2 Comparison with Map (Red-Black Tree)
- 3.3 Application in Japanese
- 3.4 Other Applications
- 4 Experiments
- 4.1 Results
- 4.2 Performance Analysis
- 5 Conclusion
- References
- Research on the Extraction of Wikipedia-Based Chinese-Khmer Named Entity Equivalents
- 1 Introduction
- 2 Analysis on the Wikipedia Page Structure
- 3 Extraction of Chinese- Khmer Named Entity Equivalences Based on the Internal Links on Wikipedia
- 4 Extraction of Chinese- Khmer Named Entity Equivalence Based on Feature Similarity
- 4.1 Characteristics of Transliteration
- 4.2 Translation Features
- 5 Experiment and Assessment
- 6 Conclusions
- References
- Bilingual Lexicon Extraction with Temporal Distributed Word Representation from Comparable Corpora
- 1 Introduction
- 2 Background: Linear Translation Transformation
- 3 Temporal Distributed Word Representation
- 4 Experiment and Results
- 4.1 Experimental Settings
- 4.2 Results
- 5 Related Works
- 6 Conclusions
- References
- Bilingually-Constrained Recursive Neural Networks with Syntactic Constraints for Hierarchical Translation Model
- 1 Introduction
- 2 Bilingually-Constrained RNN
- 2.1 The BC-RNN Model
- 2.2 The Objective Function
- 2.3 Parameter Inference
- 3 A Hierarchical Phrase-Based Translation Model that Leverages Syntactic Information
- 3.1 Feature1: The Score of the Parse Tree
- 3.2 Feature2: Hierarchical Rules with Syntactic Categories
- 4 Experiments
- 4.1 Data Preparation and Tools
- 4.2 Machine Translation Performance
- 5 Conclusion
- References
- Document-Level Machine Translation Evaluation Metrics Enhanced with Simplified Lexical Chain
- 1 Introduction
- 2 Related Work
- 3 Evaluation Data
- 4 Text Cohesion Representing by Simplified Lexical Chain
- 4.1 Simplified Lexical Chain
- 4.2 The Characteristics of Lexical-Chain Index
- 4.3 Text Cohesion Scores Based on the Matching of Lexical Chain
- 5 Experiments
- 6 Conclusion
- References
- Cross-Lingual Tense Tagging Based on Markov Tree Tagging Model
- 1 Related Work
- 2 Tense Trees
- 3 Markov Tree Tagging Model
- 3.1 Introduction to Tree Tagging Models
- 3.2 The Definition of MTTM
- 3.3 Left-Most Derivation of the Tagged Tree T[X]
- 3.4 Using Dynamic Programming to Find the Best Tagged Tree
- Use Markov Tree Tagging Model to Tag Tense Trees
- 5 Experimental Results and Discussion
- 5.1 Experimental Data
- 5.2 Experimental Results
- 6 Conclusions
- References
- Building a Large-Scale Cross-Lingual Knowledge Base from Heterogeneous Online Wikis
- 1 Introduction
- 2 Preliminaries
- 3 Semantic Data Extraction
- 4 Cross-lingual Integration
- 5 Result
- 6 Related Work
- 7 Conclusion
- References
- Refining Kazakh Word Alignment Using Simulation Modeling Methods for Statistical Machine Translation
- 1 Introduction
- 2 Description of Our Method
- 3 Evaluation
- 4 Conclusions
- References
- A Local Method for Canonical Correlation Analysis
- 1 Introduction
- 2 Related Work
- 2.1 Background
- 3 Local Linear Model
- 4 Experiment
- 4.1 Datasets
- 4.2 Results and Analysis
- 5 Conclusions
- References
- Learning to Rank Microblog Posts for Real-Time Ad-Hoc Search
- 1 Introduction
- 2 Related Work
- 3 Learning to Rank Model
- 4 Experiments
- 4.1 Data Collection and Set-up
- 4.2 Experimental Results
- 4.3 Feature Study
- 5 Conclusion and Future Work
- References
- Fuzzy-Rough Set Based Multi-labeled Emotion Intensity Analysis for Sentence, Paragraph and Document
- 1 Introduction
- 2 Related Work
- 3 Preliminary
- 4 Improving Fuzzy-Rough Set for Emotion Intensity Prediction
- 5 Experiments
- 5.1 Dataset and Evaluation Metric
- 5.2 Experiment Results
- 5.3 Discussion
- 6 Conclusion and Future Work
- References
- What Causes Different Emotion Distributions of a Hot Event? A Deep Event-Emotion Analysis System on Microblogs
- 1 Introduction
- 2 Framework of Event-Emotion Analysis System
- 2.1 Microblog Fetcher
- 2.2 Hot Event Detection
- 2.3 Emotion Predictor
- 2.4 Emotion Causation Analysis
- 2.5 Other Analysis Result Display
- 3 Experiments
- 3.1 Dataset
- 3.2 Results of the Emotion Causation Analysis
- 4 Related Work
- 5 Conclusion and Future work
- References
- Deceptive Opinion Spam Detection Using Deep Level Linguistic Features
- 1 Introduction
- 2 Related Work
- 3 Construction of Dataset
- 3.1 Truthful Dataset
- 3.2 Deceptive Dataset
- 3.3 Human Judge
- 4 Deep Features for Deceptive Spam Detection
- 4.1 Shallow Syntactic Features
- 4.2 Discourse Parsing Features
- 4.3 Sentiment Features
- 5 Experiments
- 6 Results and Discussion
- 7 Conclusion
- References
- Multi-sentence Question Segmentation and Compression for Question Answering
- 1 Introduction
- 2 Overview of Approach
- 3 Finding Linkings Between Sentences
- 4 Propagating the Linking Scores
- 4.1 Calculating the Authority
- 4.2 Propagating the Scores
- 4.3 Getting Final Segmentation
- 5 Experiments
- 6 Related Work
- 7 Conclusion
- References
- A User-Oriented Special Topic Generation System for Digital Newspaper
- 1 Introduction
- 2 Subject Heading Extraction
- 2.1 Text Preprocessing
- 2.2 Key Word Extracting Based on LDA
- 2.3 Remove Semantically Repetitive Subject Headings
- 3 Special Topic Generation
- 3.1 Compute Similarity Between News and a Topic
- 3.2 Organize and Refine the Special Topic
- 4 Experiments
- 5 Conclusion
- References
- Shared Task (Long Papers)
- Exploiting Heterogeneous Annotations for Weibo Word Segmentation and POS Tagging
- 1 Introduction
- 2 Traditional CRF for WS
- 3 Coupled CRF for WS
- 3.1 Creating Bundled Tags
- 3.2 Training Objective with Ambiguous Labelings
- 3.3 Stochastic Gradient Descent (SGD) Training with Two Datasets
- 4 Experiments
- 4.1 Datasets
- 4.2 Experiments on WS
- 4.3 Experiments on POS Tagging
- 4.4 Final Results on Test Data
- 5 Related Work
- 6 Conclusion
- References
- Entity Recognition and Linking in Chinese Search Queries
- 1 Introduction
- 2 Related Works
- 3 Methods
- 3.1 Problem Analysis and Method Procedure
- 3.2 Preprocessing
- 3.3 Entity Recognition
- 3.4 Entity Linking
- 3.5 Entity Disambiguation
- 4 Experiments
- 5 Conclusions and Outlook
- References
- BosonNLP: An Ensemble Approach for Word Segmentation and POS Tagging
- 1 Introduction
- 2 Backbone Algorithm
- 2.1 Pre-processing
- 2.2 Statistical Modeling
- 2.3 Post-processing
- 3 Ensemble Model for Open Track
- 4 Experiments
- 4.1 Closed Track Evaluation
- 4.2 Open Track Evaluation
- 5 Conclusion
- References
- Research on Open Domain Question Answering System
- 1 Introduction
- 2 Related Work
- 3 Methods
- 3.1 Question Classification
- 3.2 SPE Algorithm
- 3.3 WKE Algorithm
- 4 Experiment
- 4.1 Data Set
- 4.2 Evaluation Metric
- 4.3 Experimental Results
- 5 Conclusion
- References
- Overview of the NLPCC 2015 Shared Task: Chinese Word Segmentation and POS Tagging for Micro-blog Texts
- 1 Introduction
- 2 Data
- 2.1 Background Data
- 3 Description of the Task
- 3.1 Subtasks
- 3.2 Tracks
- 4 Participants
- 5 SubTask 1: Chinese Word Segmentation
- 5.1 Baseline Systems
- 5.2 Participant Systems
- 6 SubTask 2: Joint Chinese Word Segmentation and POS Tagging
- 7 Analysis
- 8 Conclusion
- References
- Overview of the NLPCC 2015 Shared Task: Entity Recognition and Linking in Search Queries
- 1 Introduction
- 2 Task Definition
- 3 Data
- 4 Scoring Metrics
- 5 The Teams
- 6 Results
- 7 Conclusion
- References
- Overview of the NLPCC 2015 Shared Task: Weibo-Oriented Chinese News Summarization
- 1 Task
- 2 Data
- 3 Participants
- 4 Results
- 5 Conclusion and Future Work
- References
- Overview of the NLPCC 2015 Shared Task: Open Domain QA
- 1 Background
- 2 Task Description
- 3 Benchmark Data
- 4 Auxiliary Data
- 5 Evaluation Metric
- 6 Evaluation Result
- 7 Conclusion
- Reference
- Short Task (Short Papers)
- Word Segmentation of Micro Blogs with Bagging
- 1 Introduction
- 2 System Description
- 2.1 Basic Model
- 2.2 Feature Templates
- 2.3 Bagging Model
- 2.4 Rule-Based Adaptation
- 3 Experiments
- 3.1 Effect of Preprocessing and Bagging
- 3.2 Effect of Statistic-Based Features
- 3.3 Effect of Lexicon Features
- 3.4 Final System
- 4 Conclusion
- References
- Weibo-Oriented Chinese News Summarization via Multi-feature Combination
- 1 Introduction
- 2 Related Work
- 3 System Description
- 3.1 System Architecture
- 3.2 Feature Extraction and Combination
- 3.3 Sentence Selection
- 4 Evaluation Results and Discussions
- 5 Conclusions and Future Work
- References
- Linking Entities in Chinese Queries to Knowledge Graph
- 1 Introduction
- 2 The Proposed Approach
- 2.1 Mention Identification
- 2.2 Features of Mention-Entity Pairs
- Name Length
- Priori Probability.
- Entity Relatedness
- Document Similarity.
- 2.3 Link Prediction
- 3 Evaluation Result
- 3.1 Dataset
- 3.2 Evaluation Metrics
- 3.3 Results
- 4 Related Work
- 5 Conclusion and Future Work
- References
- A Hybrid Re-ranking Method for Entity Recognition and Linking in Search Queries
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 Pre-processing
- 3.2 Named Entity Recognition
- 3.3 Candidate Entity Generation
- 3.4 Candidate Entity Re-ranking
- 4 Experiments
- 4.1 Dataset
- 4.2 Experiment Result and Analysis
- 5 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.