Semantic Web Evaluation Challenges

Name: Semantic Web Evaluation Challenges | Second SemWebEval Challenge at ESWC 2015, Portoroz, Slovenia, May 31 - June 4, 2015, Revised Selected Papers
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

Second SemWebEval Challenge at ESWC 2015, Portoroz, Slovenia, May 31 - June 4, 2015, Revised Selected Papers

Fabien Gandon Elena Cabrio Milan Stankovic Antoine Zimmermann(Editor)

Springer (Publisher)

Published on 30. September 2015

XV, 265 pages

E-Book

PDF with digital watermarking

System requirements

978-3-319-25518-7 (ISBN)

€53.49incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Preface
Organization
Contents
Open Knowledge Extraction Challenge (OKE-2015)
Open Knowledge Extraction Challenge
1 Introduction
2 Tasks
2.1 Task 1: Entity Recognition, Linking and Typing for Knowledge Base population
2.2 Task 2: Class Induction and Entity Typing for Vocabulary and Knowledge Base Enrichment
3 Training and Evaluation Datasets
3.1 Task 1
3.2 Task 2
4 Results
5 Conclusions
References
CETUS -- A Baseline Approach to Type Extraction
1 Introduction
2 Related Work
3 Pattern Extraction
3.1 Sentence Part Extraction
3.2 Grammar Construction
4 Type Extraction
4.1 Type String Extraction
4.2 Local Type Hierarchy
5 Entity Type Linking Using YAGO
6 Entity Type Linking Using FOX
7 Evaluation
7.1 OKE Challenge 2015 Task 1
7.2 OKE Challenge 2015 Task 2
8 Conclusion
References
A Hybrid Approach for Entity Recognition and Linking
1 Introduction
2 Related Work
2.1 Entity Recognition
2.2 Entity Linking
3 A Hybrid Approach for Entity Recognition and Linking
3.1 Named Entity Recognition
3.2 Named Entity Linking
3.3 Named Entity Pruning
4 System Implementation
4.1 Pipeline_1
4.2 Pipeline_2
4.3 Pipeline_3
5 Experimental Settings and Results
5.1 Statistics of the Oracle
5.2 Experimental Settings
5.3 Results on the Training Set
5.4 Comparison with Other Tools on the Training Set
5.5 Results on the Test Set
5.6 Comparison with Other Tools on the Test Set
6 Conclusion and Future Work
References
Using FRED for Named Entity Resolution, Linking and Typing for Knowledge Base Population
1 Introduction
2 FRED at Work
3 Addressing the Open Knowledge Extraction Challenge
3.1 Task 1: Named Entity Resolution, Linking and Typing for Knowledge Base Population
3.2 Task 2: Class Induction and Entity Typing for Vocabulary and Knowledge Base Enrichment
4 Results
5 Conclusions
References
Exploiting Linked Open Data to Uncover Entity Types
1 Introduction
2 Related Work
3 Methodology
3.1 Type Induction
3.2 Type Alignment
4 Evaluation
4.1 Type Induction
4.2 Type Alignment
4.3 Competition Result
5 Conclusion
References
Semantic Publishing Challenge (SemPub2015)
Semantic Publishing Challenge -- Assessing the Quality of Scientific Output by Information Extraction and Interlinking
1 Introduction: Semantic Publishing Today
2 Definition of the Challenge
3 Common Evaluation Procedures
4 Task 1: Extraction and Assessment of Workshop Proceedings Information
4.1 Motivation and Objectives
4.2 Data Source
4.3 Queries
4.4 Accepted Submissions and Winners
4.5 Lessons Learnt
5 Task 2: Extracting Contextual Information from the PDF Full Text of the Papers
5.1 Motivation and Objectives
5.2 Data Source
5.3 Queries
5.4 Accepted Submissions and Winner
5.5 Lessons Learnt
6 Task 3: Interlinking
6.1 Motivation and Objectives
6.2 Data Source
6.3 Queries
6.4 Lessons Learnt
7 Overall Lessons Learnt for Future Challenges
References
Information Extraction from Web Sources Based on Multi-aspect Content Analysis
1 Introduction
2 System Architecture
2.1 Page Rendering
2.2 Model Building
2.3 Output Dataset Generation
3 Ontological Model
3.1 Rendered Page Level
3.2 Segmentation Level
3.3 Tagging Level
3.4 Logical Tree Level
3.5 Domain Level
4 System Implementation
4.1 Layout Analysis
4.2 Generic Text Tagging
4.3 CEUR Entity Recognition
4.4 Logical Structure Construction
4.5 CEUR Index Page Processing
5 Conclusions
References
Extracting Contextual Information from Scientific Literature Using CERMINE System
1 Introduction
2 System Overview
2.1 Models and Formats
2.2 System Architecture
3 Metadata Extraction Algorithms
3.1 Geometric Structure Extraction
3.2 Content Classification
3.3 Author and Affiliation Extraction
3.4 Affiliation Parsing
3.5 References Extraction
3.6 Reference Parsing
4 Semantic Publishing Challenge of ESWC 2015
5 Conclusions and Future Work
References
Machine Learning Techniques for Automatically Extracting Contextual Information from Scientific Publications
1 Introduction
2 Unsupervised Extraction of Contiguous Text Blocks as Basic Units of a PDF
3 Supervised Classification of Author and Affiliation Meta-Data
3.1 Classification of Text Blocks
3.2 Classification of Tokens
4 Detection, Segmentation, and Tokenisation of References
4.1 Reference Line Extraction
4.2 Reference Segmentation
4.3 Reference Preprocessing
4.4 Reference Token Classification
5 Extracting Funding Information Using Named Entity Recognition
6 Conclusion and Discussion
References
MACJa: Metadata and Citations Jailbreaker
1 Introduction
2 Related Work
3 Metadata and Citations Jailbreaker
3.1 Materials and Ontologies
3.2 Methods
4 Implementation Details
4.1 Queries Q2.1 and Q2.2: Affiliations
4.2 Queries Q2.3, Q2.4 and Q2.5: Citations
4.3 Queries Q2.6, Q2.7 and Q2.8: Research Grants, Funding Agencies and EU Projects
4.4 Queries Q2.9 and Q2.10: Related and New Ontologies
5 Conclusions
References
Automatic Construction of a Semantic Knowledge Base from CEUR Workshop Proceedings
1 Introduction
2 Design
2.1 Syntactic Processing
2.2 Semantic Processing
2.3 Knowledge Base Construction
3 Implementation
3.1 Text Pre-processing
3.2 Rule-Based Extraction of Contextual Entities
3.3 Knowledge Base Population
4 Results and Discussion
5 Conclusions
References
CEUR-WS-LOD:
1 Introduction
2 System Description
3 Ontology Model
3.1 Mapping to Well-Know Ontologies
4 Overview of Dataset
4.1 Example Queries
5 Conclusion
References
Metadata Extraction from Conference Proceedings Using Template-Based Approach
1 Introduction
2 Data Model
3 Our Approach
4 Implementation
4.1 Overall Architecture
4.2 Library for Context Information Extraction from the PDF Full Text of the Papers
5 Results and Discussions
6 Conclusion
References
Semantically Annotating CEUR-WS Workshop Proceedings with RML
1 Introduction
2 Problem Statement
3 Overview of Our Approach
4 RML
4.1 Structure of an RML Mapping Document
4.2 Leveraging HTML with RML
5 Data Modeling
6 Mapping CEUR-WS from HTML to RDF
6.1 Defining the Mappings
6.2 Executing the Mappings
7 Query Evaluation
8 Tools
8.1 RML Processor
8.2 RDFUnit
8.3 The DataTank
9 Discussion and Conclusion
References
On the Automated Generation of Scholarly Publishing Linked Datasets: The Case of CEUR-WS Proceedings
1 Extract and Semantically Model Scholarly Publishing Contents
2 Turning On-line Workshop Proceedings into RDF Graphs: Overall Approach
3 Data Analysis Pipelines
3.1 Task 1: Processing CEUR-WS HTML Contents
3.2 Task 2: Mining PDF Papers
4 Modeling Workshop Data as an RDF Graph
5 Evaluating Workshop Linked Datasets by SPARQL Queries
6 Conclusions and Future Work
References
Schema-Agnostic Queries over Large-Schema Databases Challenge (SAQ-2015)
The Schema-Agnostic Queries (SAQ-2015) Semantic Web Challenge: Task Description
1 Introduction
2 Schema-Agnostic Queries
3 Challenge Description
4 Evaluation Description
4.1 Schema-Agnostic SPARQL Query
4.2 Schema-Agnostic Keyword Query
4.3 Returned Result
5 Schema-Agnostic Mappings
6 Results
7 Summary
References
UMBC_Ebiquity-SFQ: Schema Free Querying System
Abstract
1 Introduction
2 Semantic Similarity Component
3 Concept Level Association Knowledge Model (CAK Model)
4 Query Interpretation
5 Type Inference and Property Mapping
6 SPARQL Query Generation and Selection
7 System II
8 Evaluation and Discussion
9 Conclusions
References
Concept-Level Sentiment Analysis Challenge (CLSA2015)
ESWC 15 Challenge on Concept-Level Sentiment Analysis
1 Introduction
2 Related Work
3 Tasks, Datasets and Evaluation Measures
3.1 Task 1: Polarity Detection
3.2 Task 2: Aspect-Based Sentiment Analysis
3.3 Task 3: Frame Entities Identification
3.4 Task 4: The Most Innovative Approach
4 Submitted Systems
5 Results
5.1 Task 1
5.2 Task 3
5.3 The Most Innovative Approach Task
6 Conclusions
References
The Benefit of Concept-Based Features for Sentiment Analysis
1 Introduction
2 Related Work
3 Sentence Polarity Classification
3.1 Data
3.2 Evaluation
4 Aspect Detection
4.1 Data
4.2 Evaluation
5 Aspect Polarity Classification
5.1 Evaluation
6 Conclusion
References
An Information Retrieval-Based System for Multi-domain Sentiment Analysis
1 Introduction
2 Related Work
3 The SHELLFBK System
3.1 Indexes Construction
3.2 Domain and Polarity Computation
4 In-Vitro Evaluation and Challenge Results
4.1 In-Vitro Evaluation
4.2 Participation at ESWC 2015 Challenge on Semantic Sentiment Analysis
5 Conclusion
References
Detecting Sentiment Polarities with Sentilo
1 Introduction
2 Description of Sentilo
3 Addressing the Concept-Level Sentiment Analysis Challenge
4 Results
5 Conclusions
References
Supervised Opinion Frames Detection with RAID
1 Introduction
2 Related Work
2.1 Approaches for Opinion Frames Extraction
2.2 Datasets Annotated with Opinion Frames
3 The RAID Pipeline
3.1 Pre-processing
3.2 Extraction of Opinion Expressions
3.3 Extraction of Opinion Holders/Targets
3.4 Polarity Classification
4 Evaluation
5 Conclusions and Future Work
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Semantic Web Evaluation Challenges

Description

More details

Other editions

Additional editions

Content

System requirements