
Experimental IR Meets Multilinguality, Multimodality, and Interaction
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
The 10 full papers and 8 short papers presented together with 5 best of the labs papers were carefully reviewed and selected from 36 submissions. In addition to these talks, this volume contains the results of 7 benchmarking labs reporting their year long activities in overview talks and lab sessions. The papers address all aspects of information access in any modality and language and cover a broad rangeof topics in the fields of multilingual and multimodal information access evaluation.
More details
Other editions
Additional editions

Content
- Intro
- Preface
- Organization
- Contents
- Cross Language Information Retrieval and Evaluation
- Show Me How to Tie a Tie: Evaluation of Cross-Lingual Video Retrieval
- 1 Introduction
- 2 Related Work
- 3 Data
- 4 Results
- 4.1 Manual Analysis of 1,000 Queries
- 4.2 Machine Translation
- 4.3 Video Retrieval Evaluation
- 5 Conclusions
- References
- The CLEF Monolingual Grid of Points
- 1 Introduction
- 2 Overview of CLEF Monolingual Tasks
- 3 Grid of Points
- 4 Analysis of the Grid of Points
- 5 Final Remarks
- References
- A Test Collection for Research on Depression and Language Use
- 1 Introduction
- 2 Building a Textual Collection for Depression
- 2.1 Selection of Data Source
- 2.2 Depression Group
- 2.3 Control Group
- 2.4 Texts Extracted
- 2.5 Resulting Collection
- 3 Early Prediction Task
- 3.1 Evaluation Metric
- 4 Baseline Experiments
- 5 Conclusions
- References
- Assessors Agreement: A Case Study Across Assessor Type, Payment Levels, Query Variations and Relevance Dimensions
- 1 Introduction
- 2 Related Work
- 3 Data
- 4 Agreements for Topical Relevance Assessments
- 4.1 Inter-assessor Agreement Among Paid Assessors
- 4.2 Influence of Assessor Type and Payment Level
- 4.3 Assessor Agreement Across Query Variations
- 5 Agreements for Understandability Assessments
- 5.1 Inter-assessor Agreement Among Paid Assessors
- 5.2 Influence of Assessor Type and Payment Level
- 5.3 Assessors Agreement Across Query Variations
- 6 Conclusions
- References
- Reranking Hypotheses of Machine-Translated Queries for Cross-Lingual Information Retrieval
- 1 Introduction
- 2 Related Work
- 3 Experimental Setting
- 3.1 Data
- 3.2 Retrieval System
- 3.3 Translation System
- 4 Method
- 5 Experiments and Results
- 5.1 Monolingual Performance
- 5.2 Baseline
- 5.3 Oracle Results
- 5.4 n-best List Merging
- 5.5 Reranking
- 6 Conclusions
- References
- Classification, Profiling and Suggestion
- Two-Way Parsimonious Classification Models for Evolving Hierarchies
- 1 Introduction
- 2 Related Work
- 3 Hierarchical Significant Words Language Models
- 3.1 Model Parsimonization
- 3.2 Estimating HSWLM
- 4 HSWLM for Evolving Hierarchies
- 4.1 Data Collection and Experimental Settings
- 4.2 Classification Across Periods
- 4.3 Invariance of Models
- 5 Conclusions
- References
- Effects of Language and Terminology on the Usage of Health Query Suggestions
- 1 Introduction
- 2 Related Work
- 3 Suggestion Tool
- 4 Experiment
- 5 Data Analysis
- 6 Use of Suggestions
- 6.1 Analysis by Language
- 6.2 Analysis by Terminology
- 7 Discussion
- 8 Conclusion
- References
- Predicting Contextually Appropriate Venues in Location-Based Social Networks
- 1 Introduction
- 2 Related Work
- 3 Problem Statement
- 4 Contextual Aspect Features
- 4.1 Categorical and Temporal Features
- 4.2 Term-Based Features
- 5 Venue Dimension Classification Evaluation
- 5.1 Crowdsourcing Venue Annotations
- 5.2 Experimental Setup
- 5.3 Experimental Results
- 6 Ranking Contextually Appropriate Venues
- 6.1 Experimental Setup
- 6.2 Experimental Results
- 7 Conclusions
- References
- I, Me, Mine: The Role of Personal Phrases in Author Profiling
- 1 Introduction
- 2 Related Work
- 3 Social Media Datasets
- 4 Experimental Methodology
- 4.1 Filtering Process
- 4.2 Classification Process
- 5 Results and Discussion
- 5.1 Experiment 1: The Relevance of Personal Phrases for AP
- 5.2 Experiment 2: The Added Value of Plural Personal Phrases
- 5.3 Experiment 3: Content and Style Information in Personal Phrases
- 5.4 Experiment 4: Personal Phrases in Different Social Media
- 6 Conclusions and Future Work
- References
- Improving Profiles of Weakly-Engaged Users
- 1 Introduction
- 2 Related Work
- 3 Proposed Model
- 4 Experimentation
- 4.1 Dataset
- 4.2 User Profile Completion
- 4.3 Recommendations
- 5 Conclusions
- References
- Best of CLEF 2015 Labs
- Kronecker Decomposition for Image Classification
- 1 Introduction
- 2 Image Feature Generation via Decomposition
- 2.1 Reordering of the Matrix
- 2.2 Kronecker Decomposition as SVD
- 3 Learning Task
- 4 Optimization Model
- 4.1 The ``Classical'' Scheme of Support Vector Machine (SVM)
- 4.2 Reinterpretation of the Normal Vector w
- 4.3 Dual Problem
- 4.4 Prediction
- 4.5 Hierarchy Learning
- 4.6 Input and Output Kernels
- 5 Experimental Evaluation
- 5.1 Methods
- 5.2 ImageCLEF Multi-label Classification
- 5.3 Pascal and Flickr
- 6 Discussion
- 7 Conclusions
- References
- A Two-Step Retrieval Method for Image Captioning
- 1 Introduction
- 2 Related Work
- 3 A Two-Step Retrieval Method for Image Captioning
- 3.1 Feature Extraction
- 3.2 Multimodal Indexing
- 3.3 Step 1: Word-Retrieval (WR)
- 3.4 Step 2: Caption-Retrieval (CR)
- 4 Experimental Settings
- 4.1 Settings for Query Formulation (from WR Step to CR Step)
- 5 Experimental Results
- 5.1 Auto-illustration
- 6 Conclusions
- References
- Concept Recognition in French Biomedical Text Using Automatic Translation
- 1 Introduction
- 2 Methods
- 2.1 Corpora
- 2.2 Term Translation
- 2.3 Terminologies
- 2.4 Entity Recognition and Entity Normalization
- 2.5 Normalization Based on Gold-Standard Entity Recognition
- 3 Results
- 3.1 Performance on the Quaero Training Set
- 3.2 Performance on the Quaero Test Data
- 4 Discussion
- References
- A Product Feature-Based User-Centric Ranking Model for E-Commerce Search
- 1 Introduction
- 2 Related Work
- 3 Product Feature-Based User-Centric Ranking Model
- 3.1 The Model
- 3.2 Experimental Evaluation
- 4 Understanding the Effect of the User Engagement Factor in Product Search
- 5 Discussion and Conclusion
- References
- Random Performance Differences Between Online Recommender System Algorithms
- 1 Introduction
- 2 Approach
- 3 Experiments
- 4 Results and Analysis
- 5 Causes of Performance Differences
- 5.1 Operation Causes
- 5.2 User-Item Causes
- 5.3 Random Causes
- 5.4 Overlap in Performance
- 6 Conclusion
- References
- Short Papers and Posters
- SS4MCT: A Statistical Stemmer for Morphologically Complex Texts
- 1 Introduction
- 2 SS4MCT: A Statistical Stemmer for Morphologically Complex Texts
- 2.1 Transformation Rules
- 2.2 Probability of the Rules
- 2.3 How to Evaluate the Algorithm
- 3 Experiments
- 3.1 Experimental Setup
- 3.2 Comparing Different Morphological Processing Methods
- 4 Conclusion and Future Works
- References
- Index-Based Semantic Tagging for Efficient Query Interpretation
- 1 Introduction
- 2 Reference Work
- 3 Data Characterization
- 4 Semantic Tagging in Queries
- 5 Evaluation
- 5.1 Comparing Query Analysis Time for the Retrieval Strategies
- 5.2 Influence of N over the Speedup
- 6 Conclusions
- References
- A Gamified Approach to Relevance Judgement
- 1 Introduction
- 2 DocMiner: Gamified Relevance Assessment
- 2.1 Game Properties
- 3 User Study Evaluation
- 4 Conclusions and Future Work
- References
- Evaluating Categorisation in Real Life -- An Argument Against Simple but Impractical Metrics
- 1 The Use Case: Practical Cold Start Text Categorisation
- 2 Keyword-Based Categorisation
- 2.1 The F-Score Evaluation Metric
- 2.2 The R@P Curve
- 2.3 P & R Matrix --- A More Expressive Representation
- 3 Conclusions
- References
- How Relevant is the Long Tail?
- 1 Introduction
- 2 Materials and Methods
- 3 Results
- 4 Discussion and Conclusion
- References
- Towards an Understanding of Transactional Tasks
- 1 Introduction
- 2 Experimental Setup
- 2.1 Setup and Design
- 2.2 Tasks
- 3 Results and Analysis
- 3.1 Perceived Difficulty
- 3.2 Search Engine Support
- 3.3 Task Completion Patterns
- 4 Conclusions
- References
- Health Suggestions: A Chrome Extension to Help Laypersons Search for Health Information
- 1 Introduction
- 2 Related Work
- 3 Health Suggestions
- 4 Experiment
- 4.1 Setup
- 4.2 Results
- 5 Conclusions and Future Work
- References
- Brazilian Social Mood: The Political Dimension of Emotion
- Abstract
- 1 Introduction
- 2 Related Work
- 3 The Proposed Approach
- 4 Evaluation and Results
- 5 Conclusions and Future Works
- References
- CLEF 2016 Labs Overviews
- Overview of the CLEF eHealth Evaluation Lab 2016
- 1 Introduction
- 2 Materials and Methods
- 2.1 Text Documents
- 2.2 Human Annotations, Queries, and Relevance Assessments
- 2.3 Evaluation Methods
- 3 Results
- 4 Conclusions
- References
- General Overview of ImageCLEF at the CLEF 2016 Labs
- 1 Introduction
- 2 Overview of Tasks and Participation
- 3 The Medical Task
- 3.1 Past Editions
- 3.2 Objectives and Subtasks for the 2016 Edition
- 3.3 Participation and Results
- 4 The Image Annotation Task
- 4.1 Past Editions
- 4.2 Objective and Task for the 2016 Edition
- 4.3 Participation and Results
- 5 The Handwritten Retrieval Task
- 5.1 Previous Work
- 5.2 Objective and Task for the 2016 Edition
- 5.3 Participation and Results
- 6 Conclusions
- References
- LifeCLEF 2016: Multimedia Life Species Identification Challenges
- 1 LifeCLEF Lab Overview
- 2 Task1: PlantCLEF
- 2.1 Dataset and Evaluation Protocol
- 2.2 Participants and Results
- 3 Task2: BirdCLEF
- 3.1 Data and Task Description
- 3.2 Participants and Results
- 4 Task3: SeaCLEF
- 4.1 Coral Reef Species Identification in Underwater Videos
- 4.2 Individual Humpback Whale Identification
- 5 Conclusions and Perspectives
- References
- Overview of NewsREEL'16: Multi-dimensional Evaluation of Real-Time Stream-Recommendation Algorithms
- 1 Introduction
- 2 Related Work
- 2.1 Benchmarking in Static Environments
- 2.2 Benchmarking in Dynamic Environments
- 3 Problem Description
- 3.1 Task 1: Benchmark News Recommendations in a Living Lab
- 3.2 Task 2: Benchmark News Recommendations in a Simulated Environment
- 3.3 Summary
- 4 Multi-dimensional Evaluation Online and Offline
- 4.1 Online Evaluation Methods
- 4.2 Offline Evaluation Methods
- 4.3 Discussion
- 5 The Participant Perspective
- 5.1 Opportunities
- 5.2 Validity and Fairness
- 5.3 Discussion
- 6 Evaluation Results
- 6.1 Task 1: Online Competition
- 6.2 Task 2: Offline Evaluation
- 6.3 Comparing Online and Offline
- 6.4 Participation
- 6.5 Discussion
- 7 Conclusion and Outlook
- References
- Overview of PAN'16
- 1 Introduction
- 2 Author Identification
- 2.1 Author Clustering
- 2.2 Author Diarization
- 3 Author Profiling
- 4 Author Obfuscation
- 5 Conclusions
- References
- Overview of the CLEF 2016 Social Book Search Lab
- 1 Introduction
- 2 Participating Organisations
- 3 The Amazon/LibraryThing Corpus
- 4 Suggestion Track
- 4.1 Track Goals and Background
- 4.2 Information Needs
- 4.3 Evaluation
- 5 Interactive Track
- 5.1 User Tasks
- 5.2 Experiment Structure
- 5.3 System and Interfaces
- 5.4 Participants
- 5.5 Procedure
- 5.6 Results
- 6 Mining Track
- 6.1 Track Goals and Background
- 6.2 Track Setup
- 6.3 Evaluation
- 6.4 Results
- 7 Conclusions and Plans
- References
- Overview of the CLEF 2016 Cultural Micro-blog Contextualization Workshop
- 1 Introduction
- 1.1 Context
- 1.2 Aims of the Workshop
- 2 Use Case Scenario
- 3 Datasets
- 3.1 Micro-blog Posts Collection
- 3.2 Linked Web Pages
- 3.3 Wikipedia Crawl
- 4 Micro-blog Corpus Insights
- 5 Experimented Tasks
- 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.