Intro
Preface
Organization
Contents
Part II
Contents
Part I
Information Extraction
TaskGenie: Crowd-Powered Task Generation for Struggling Search
1 Introduction
2 Related Literature
3 Task Generation Framework
3.1 Intuition and Method
3.2 TaskGenie
A Crowd-Powered Platform
3.3 System Implementation
4 Study Design
4.1 Task Generation
4.2 Task Evaluation
5 Task Evaluation Results
5.1 Topical Characteristics
5.2 Query Characteristics
5.3 Clicks Characteristics
5.4 Task Difficulty Analysis
5.5 Publicly Released Task Set

6 Discussion
7 Conclusions and Future Work
References
Search Engine Similarity Analysis: A Combined Content and Rankings Approach
1 Introduction
2 Background and Related Work
3 The Metric
3.1 Problem Formulation
3.2 Starting Point
3.3 Calculation of Penalties
3.4 Similarity Boosting
3.5 The Metric T
3.6 Comparison with Other Metrics
4 Evaluation
4.1 Dataset
4.2 RQ1: Similarity of Search Engines
4.3 RQ2: Consistency of Search Engines
4.4 RQ3: Impact of Snippets, Titles, and Transpositions

4.5 RQ4: Search Engine Similarity in Different Search Services
4.6 RQ5: Comparison with Other Approaches
4.7 Threats to Validity
5 Conclusions and Discussion
References
Evaluating Similarity Measures for Dataset Search
1 Introduction
2 Preliminaries
2.1 Similarity Measures
2.2 Evaluation Measures for Information Retrieval
3 Similarity Measure and Evaluation Measure
3.1 Similarity Between Sets of Concepts
3.2 Evaluation Measure
4 Experiments and Results
4.1 Data Selection
4.2 Extract Terms from Query or Dataset
4.3 Similarity Experiments

4.4 Experimental Results
5 Discussion and Conclusion
References
Towards Efficient Retrieval of Top-k Entities in Systems of Engagement
1 Introduction
2 Context of the Problem
3 Top-K Query Processing Schemes
4 The HI-TREE Index for Computing Relevance Scores
5 Performance Evaluation
6 Conclusion
References
A Compare-Aggregate Model with External Knowledge for Query-Focused Summarization
1 Introduction
2 Background
2.1 Compare-Aggregate Model
2.2 External Knowledge
3 Approach
3.1 Problem Formulation
3.2 Sentence Relevance Scorer

3.3 Sentence Selection
4 Experiments and Results
4.1 Experimental Setup
4.2 Baselines
4.3 Evaluation Results
5 Related Work
5.1 Unsupervised Methods
5.2 Supervised Methods
5.3 Semi-supervised Methods
6 Conclusion and Future Work
References
An Active Learning Based Hybrid Neural Network for Joint Information Extraction
1 Introduction
2 Preliminaries
2.1 Problem Formulation
2.2 Related Work
3 The Framework
4 Joint IE Model
4.1 Word Embedding Layer
4.2 Bidirectional LSTM Encoding Layer
4.3 Graph Attention Network Layer