Semantic Technology: 9th Joint International Conference, JIST 2019, Hangzhou, China, November 25–27, 2019, Proceedings (Information Systems and Applications, incl. Internet/Web, and HCI Book 12032)

This book constitutes the thoroughly refereed proceedings of the 9th Joint International Semantic Technology Conference, JIST 2019, held in Hangzhou, China, in November 2019. The 24 full papers presented were carefully reviewed and selected from 70 submissions. They present applications of semantic technologies, theoretical results, new algorithms and tools to facilitate the adoption of semantic technologies and are organized in topical sections on knowledge graphs; data management; question answering and NLP; ontology and reasoning; government open data; and semantic web for life sciences. Preface Organization Contents Incorporating Term Definitions for Taxonomic Relation Identification 1 Introduction 2 Related Work 3 The Proposed Method 3.1 The Baseline System 3.2 Our Proposed Model 4 Experiments and Analysis 4.1 Dataset 4.2 Experimental Settings 4.3 Performance on Specific and Open Domain Datasets 4.4 Error Analysis 5 Conclusion References Report on the First Knowledge Graph Reasoning Challenge 2018 1 Background and Goal of the Challenge 2 Knowledge Graph Construction 2.1 Process of Holding the Challenge 2.2 Details of the Schema 3 Approach for Estimation and Reasoning Techniques 3.1 Submission of NRI 3.2 Submission of Team Kamikotanaka 411, Fujitsu 3.3 Submission of Team FLL-ML, Fujitsu 3.4 Submission of Nagoya Institute of Technology 4 Evaluation 4.1 Basic Information 4.2 Expert Evaluation 4.3 General Examination 4.4 Evaluation Results 5 Related Work 6 Conclusion and Future Work References Violence Identification in Social Media 1 Introduction 2 Background 3 Methodology 3.1 Knowledge Base Construction Stage 3.2 Social Media Data Collection 3.3 Knowledge Processing Stage 3.4 Violence Domain Identification 4 Evaluation 4.1 Discussion of the Evaluation 5 Conclusions References Event-Oriented Wiki Document Generation 1 Introduction 2 Related Work 3 Method 3.1 Topic Template Induction 3.2 Topic Summary Generation 4 Dataset Construction 5 Experiments 5.1 Dataset 5.2 Topic Template Induction 5.3 Topic Summary Generation 6 Conclusion References A Linked Data Model-View-* Approach for Decoupled Client-Server Applications 1 Introduction 2 Related Work 3 Preliminaries 3.1 Model-View-* Patterns 3.2 Entity-Component-Attribute Data Model 3.3 Linked Data Platform 3.4 Richardson Maturity Model 4 Server-Side Linked Data Generation 4.1 Linked Data Representation of Runtime Data 4.2 RESTful Operations on Runtime Data: 4.3 Realtime Subscription Pub/Sub: 5 Client-Side Data Access Layer Generation 6 Discussion 6.1 MVPVM Pattern in Resulting Architecture 6.2 RMM-Compliance of Server Interfaces 7 Conclusion and Future Work References JECI: A Joint Knowledge Graph Embedding Model for Concepts and Instances 1 Introduction 2 Preliminaries 3 JECI Model 3.1 Hierarchical Tree Generator 3.2 Context Vector Generator 3.3 Embeddings Learner 4 Experiments 4.1 Datasets 4.2 Link Prediction 4.3 Triple Classification 4.4 Limitations 5 Conclusion References Enhanced Entity Mention Recognition and Disambiguation Technologies for Chinese Knowledge Base Q&A 1 Introduction 2 Related Work 2.1 Mention Recognition Research 2.2 Entity Disambiguation Research 3 Topic Entity Mention Recognition 3.1 Algorithm Description 3.2 Feature Extraction 4 Entity Disambiguation 4.1 Entity Disambiguation Framework 4.2 Question Similarity Calculation 5 Experiments 5.1 Data Sets 5.2 Evaluation Metric 5.3 Topic Entity Mention Recognition 5.4 Entity Disambiguation 5.5 Knowledge Base Q&A Overall Experiment 6 Conclusion References Dispute Generation in Law Documents via Joint Context and Topic Attention 1 Introduction 2 Related Work 2.1 Legal-Related Work 2.2 Text Generation 3 Problem Formulation 4 Model 5 Experiments 5.1 Dataset 5.2 Implementation Details 5.3 Baseline Models 5.4 Evaluation Metrics 5.5 Evaluation Results 5.6 Case Study 6 Conclusions and Future Work References Richpedia: A Comprehensive Multi-modal Knowledge Graph 1 Introduction 2 Richpedia Construction 2.1 Data Collection for Richpedia 2.2 Richpedia Images Processing 2.3 Richpedia Relation Discovery 3 Ontology 4 Statistics and Evaluation 5 Use-Cases 6 Related Work 7 Conclusion and Future Work References DSEL: A Domain-Specific Entity Linking System 1 Introduction 2 Definition and Framework 2.1 Problem Definition 2.2 The Classical Pipeline of an Entity Linking System 2.3 Our System 3 Our Approach 3.1 Domain Generation 3.2 Mention Parsing and Candidate Generation 3.3 Domain NCEL Model 4 Experiments 4.1 Dataset 4.2 Domain Generate 4.3 Model Validation 4.4 Category Embedding 5 System Implementation 6 Related Work 7 Conclusion References Exploring the Generalization of Knowledge Graph Embedding 1 Introduction 2 Related Work 2.1 Analysis of Translation-Based Knowledge Graph Embedding Models 2.2 Analysis of Generalization in Deep Learning 3 Problem Definition 4 Analytical Methods 4.1 Measurement of Generalization Error 4.2 Measurement of Empirical Error 5 Experiment 5.1 Experimental Design 5.2 Datasets and Experimental Settings 5.3 Experimental Results 5.4 Discussion 6 Conclusion and Future Work References Incorporating Instance Correlations in Distantly Supervised Relation Extraction 1 Introduction 2 Related Work 2.1 Distantly Supervised Relation Extraction 2.2 Graph Convolution Networks 3 Our Proposed Model 3.1 Graph Construction 3.2 Graph Convolution Layer 3.3 Relation Classifier 3.4 Model Training 4 Experiments 4.1 Dataset 4.2 Baselines 4.3 Evaluation Metrics 4.4 Experiment Settings 4.5 Precision-Recall Curve on both Datasets 4.6 P@N Evaluation 4.7 Results on GIDS Dataset 4.8 Effect of Instance Number 5 Conclusion References A Physical Embedding Model for Knowledge Graphs 1 Introduction 2 Related Work 3 Preliminaries and Notation 3.1 Knowledge Graph 3.2 Hooke's Law 3.3 Positive Pointwise Mutual Information 4 Pyke 4.1 Intuition 4.2 Approach 5 Complexity Analysis 5.1 Space Complexity 5.2 Time Complexity 6 Evaluation 6.1 Experimental Setup 6.2 Results 7 Conclusion References Iterative Visual Relationship Detection via Commonsense Knowledge Graph 1 Introduction 2 Preliminary 2.1 Commonsense Knowledge Graph 2.2 Bi-directional Recurrent Neural Network 3 Method 3.1 Feature Module 3.2 Commonsense Knowledge Module 3.3 Iteration 3.4 Attention 3.5 Training 4 Experiments 4.1 Datasets 4.2 Experimental Settings 4.3 Comparative Results 5 Conclusion and Future Work References A Dynamic and Informative Intelligent Survey System Based on Knowledge Graph 1 Introduction 2 Background 2.1 Knowledge Graph 2.2 Linguistic Background 3 Requirements for an Informative and Dynamic System 4 Knowledge Graph and Algorithms 4.1 Knowledge Graph 4.2 Algorithms 5 Case Studies and Evaluations 5.1 Hypotheses 5.2 Case Study 1 5.3 Case Study 2 6 Related Work 6.1 Intelligent Surveys 6.2 Psycholinguistic Surveys 6.3 Commercial Surveys 6.4 Adaptive Questionnaires 7 Conclusion References CICO: Chemically Induced Carcinogenesis Ontology 1 Introduction 1.1 Background 1.2 Overview of Chemically Induced Carcinogenesis Ontology 2 Related Work 3 Dataset of Chemically Induced Carcinogenesis Experiments 4 Methodology for Ontology Development 4.1 Ontology Description 4.2 Scenario 4.3 Mapping to Heterogeneous Classes 5 The CICO Resource in Use 6 Conclusion and Future Work References Retrofitting Soft Rules for Knowledge Representation Learning 1 Introduction 2 Related Work 3 Retrofitting Soft Rules for Learning Knowledge Representation 3.1 Learning Resources 3.2 Triple Modeling 3.3 Distilling Soft Rules into Knowledge Representation 3.4 Retrofitting Soft Rules 3.5 Training Procedure 4 Experiments 4.1 Experimental Settings 4.2 Link Prediction 4.3 Triple Classification 5 Detailed Analysis 5.1 Confidences of Soft Rules 5.2 Failure Analysis 5.3 Learning Speed 6 Conclusions References Entity Synonym Discovery via Multiple Attentions 1 Introduction 2 Related Work 3 Problem Formulation 4 The SynMine Framework 4.1 Distant Supervision Acquisition 4.2 Global Attention Module 4.3 Piecewise CNN with Global and Local Attentions 5 Experiments 5.1 Experimental Setup 5.2 Experimental Results 5.3 Case Studies and Discussion 6 Conclusion References Towards Association Rule-Based Complex Ontology Alignment 1 Introduction 2 Related Work 3 Background 4 Association Rule-Based Alignment Algorithm 4.1 Data Preparation 4.2 Association Rule and Alignment Generation 5 Evaluation 5.1 Simple Alignment Evaluation 5.2 Complex Alignment Evaluation 6 Conclusion References Autonomous RDF Stream Processing for IoT Edge Devices 1 Introduction 2 Continuous Federation with Autonomous RSP 2.1 Preliminaries: RDF Stream Processing with CQELS-QL 2.2 Dynamic Subscription and Discovery for Autonomous RSP Engines 2.3 Continuous Query Federation Mechanism 3 Design and Implementation 4 Evaluation and Analyses 4.1 Evaluation Setup 4.2 Experiments 4.3 Results and Discussions 4.4 Follow-Up Challenges 5 Related Work 6 Conclusion References Certain Answers to a sparql Query over a Knowledge Base 1 Introduction 2 Preliminaries 2.1 rdf and sparql 2.2 Description Logic KB, UCQs and Certain Answers 3 Querying a DL KB with sparql: Existing Semantics 4 Requirements 5 Semantics 5.1 sparql Entailment Regimes 5.2 Canonical Answers 5.3 Maximal Admissible Canonical Answers 6 Complexity 7 Conclusion and Perspectives References External Knowledge-Based Weakly Supervised Learning Approach on Chinese Clinical Named Entity Recognition 1 Introduction 2 Base Model 3 External Knowledge-Based Weakly Supervised Learning CNER 3.1 External Knowledge Acquisition 3.2 Weakly Supervised Learning Data Enhancement 3.3 CNER Model Training Architecture 4 Experiments 4.1 Dataset 4.2 Experimental Settings 4.3 Overall Experimental Results 4.4 Experiment Analysis 5 Related Work 5.1 Clinical Named Entity Recognition 5.2 Weakly Supervised Learning 6 Conclusion References Metadata Application Profile Provenance with Extensible Authoring Format and PAV Ontology 1 Introduction 1.1 Application Profile Expression Formats 1.2 Current Status and Availability of Application Profiles 1.3 Challenges in Application Profile Development 1.4 Yet Another Metadata Application Profile (YAMA) as an Application Profile Authoring Format 2 Related Work 3 Methodology 3.1 Actionable Changesets as Timeline of MAP 3.2 PAV Ontology as a Means of MAP Provenance 4 Validation 5 Limitations and Future Work 6 Conclusion References An Ontology-Based Development of Activity Knowledge and System Design 1 Introduction 1.1 Motivations and Requirements 1.2 Contributions 2 Related Work 2.1 Methodologies 2.2 Knowledge Representations 3 Development of Domain Knowledge 3.1 Activity Knowledge 3.2 Domain Ontology: GRO 3.3 Ontology-Based Development of Activity Knowledge 4 System Design 4.1 Architecture 4.2 Functions 4.3 Term Control 5 Conclusion References Author Index