Information Retrieval Technology
7th Asia Information Retrieval Societies Conference, AIRS 2011, Dubai, United Arab Emirates, December 18-20, 2011, Proceedings
Published on 14. December 2011
XV, 626 pages
978-3-642-25631-8 (ISBN)
Description
This book constitutes the refereed proceedings of the 7th Asia Information Retrieval Societies Conference AIRS 2011, held in Dubai, United Arab Emirates, in December 2011.
The 31 revised full papers and 25 revised poster papers presented were carefully reviewed and selected from 132 submissions. All current aspects of information retrieval - in theory and practice - are addressed; the papers are organized in topical sections on information retrieval models and theories; information retrieval applications and multimedia information retrieval; user study, information retrieval evaluation and interactive information retrieval; Web information retrieval, scalability and adversarial information retrieval; machine learning for information retrieval; natural language processing for information retrieval; arabic script text processing and retrieval.
The 31 revised full papers and 25 revised poster papers presented were carefully reviewed and selected from 132 submissions. All current aspects of information retrieval - in theory and practice - are addressed; the papers are organized in topical sections on information retrieval models and theories; information retrieval applications and multimedia information retrieval; user study, information retrieval evaluation and interactive information retrieval; Web information retrieval, scalability and adversarial information retrieval; machine learning for information retrieval; natural language processing for information retrieval; arabic script text processing and retrieval.
More details
Other editions
Additional editions
Mohamed Vall Mohamed Salem | Khaled Shaalan | Farhad Oroumchian
Information Retrieval Technology
7th Asia Information Retrieval Societies Conference, AIRS 2011, Dubai, United Arab Emirates, December 18-20, 2011, Proceedings
Book
12/2011
Springer
€53.49
Shipment within 7-9 days
Content
- Intro
- Title page
- Preface
- Organization
- Table of Contents
- Information Retrieval Models and Theories
- Query-Dependent Rank Aggregation with Local Models
- Introduction
- Query Dependent Ranking
- Framework
- Three Approaches
- Time Complexity
- Experimental Results
- Dataset and Parameter Selection
- Performance Comparison with LETOR4 Baselines
- Analysis of Selecting the Best Local Model
- Analysis in Terms of Query Difficulty
- Related Work
- Conclusion
- References
- On Modeling Rank-Independent Risk in Estimating Probability of Relevance
- Introduction
- Literature Review
- Rank-Independent Risk Modeling
- Rank-Equivalent LM Approaches
- Difference between the Two Rank-Equivalent Estimations
- Entropy-Based Risk Measurement
- Powers-Based Risk Management (PRM) Method
- Application
- Empirical Evaluation
- Evaluation Configuration
- Evaluation on Risk Management Method for PRF Task
- Evaluation on Risk Management Method for RF Task
- Conclusions and Future Work
- Appendix A: Proof for Proposition 1
- References
- Measuring the Ability of Score Distributions to Model Relevance
- Introduction
- Related Research
- Related Work
- Contributions
- Models
- Assumptions and Restrictions
- Mixture Distributions
- Inferring Average Precision
- Mixture Performance
- Goodness-of-Fit, Correlation, and RMSE
- Comparative Analysis
- Recall-Fallout Convexity Analysis
- Locating Points of Non-convexity
- Empirical Results and Discussion
- Conclusion
- References
- Cross-Language Information Retrieval with Latent Topic Models Trained on a Comparable Corpus
- Introduction
- Related Work
- Bilingual LDA
- LDA-Based CLIR
- LDA-only CLIR Model
- LDA-Unigram CLIR Model
- Experimental Setup
- Training Collections
- Test Collections
- Results and Discussion
- Comparison with Baseline Systems
- Comparison of Our CLIR Models
- Conclusions and Future Work
- References
- Construct Weak Ranking Functions for Learning Linear Ranking Function
- Introduction
- Related Work
- Construct the Weak Ranking Functions from the Ranking Features
- Normalization Methods
- Normalization Selection Method
- Experiment
- Experimental Setting
- Dataset Statistics Information
- Experiments Results
- Conclusion and Future Work
- References
- Is Simhash Achilles?
- Introduction
- Spatial Data Analysis
- Support Vector Data Description
- Accurate Spatial Data Analysis
- Spatial Data Analysis Based Simhash (SDA-Simhash)
- Simhash
- SDA-Simhash
- Experiments
- Evaluation Criterions
- A Toy Case
- Real-World Data
- Related Work
- Conclusions
- References
- XML Information Retrieval through Tree Edit Distance and Structural Summaries
- Introduction
- Related Works
- Structural Similarities between Trees
- Semi-structured Information Retrieval
- Tree-Edit Distance for Structural Document-Query Matching
- Content Relevance Score Evaluation
- Extracting and Summarizing Subtrees
- Structure Relevance Score Evaluation
- Final Combination
- Experiments and Evaluation
- INEX Collection
- Experiments
- Conclusions and Future Work
- References
- An Empirical Study of SLDA for Information Retrieval
- Introduction
- Related Work
- SLDA Modeling Framework
- SLDA Model
- Relevant Metrics
- Experiments
- Dataset and Experiment Setup
- Comparison of Relevant Metrics
- Parameter Settings
- Comparison with Language Model
- Conclusions and Future Work
- References
- Learning to Rank by Optimizing Expected Reciprocal Rank
- Introduction
- Ranking Evaluation Criteria
- Expected Reciprocal Rank (ERR)
- Optimize ERR Metric Using Structural SVMs
- Structural SVMs
- Optimize ERR Metric Using Structural SVMs
- Experiments
- Experiment on OHSUMED Data
- Experiment on TD2003 Data
- Experimental Analysis
- Conclusion
- References
- Information Retrieval Applications and Multimedia Information Retrieval
- Information Retrieval Strategies for Digitized Handwritten Medieval Documents
- Introduction
- Related Work
- Evaluation Corpora and Methodology
- Handwritten Recognition
- The Generation of Various Evaluation Corpora
- Known-Item Query Generation
- Indexing Strategies and Retrieval Models
- Evaluation
- Evaluation of the Recognition Corpora
- Selected Query-by-Query Analyses
- Conclusion
- References
- Query Phrase Expansion Using Wikipedia in Patent Class Search
- Introduction
- Related Work
- Proposed Method
- Page Similarity
- Query Term Extraction and Expansion
- Patent Retrieval
- Experiments
- Baseline Queries
- Relevance Model
- WordNet
- Wikipedia
- Proposed Method
- Phrase Weight Balance
- Conclusions and Future Works
- References
- Increasing Broadband Subscriptions for Telecom Carriers through Mobile Advertising
- Introduction
- Related work
- Consumer Behavior and Personalized Advertising
- Web Contextual Advertising
- Mobile Advertising
- Design Methodology
- Addresses the 3 Key Issues for Mobile Advertising
- System Architecture
- Personalized Ad Matching
- Mobile Ad Collector
- Ad-crawler Platform
- Ad Feature Extraction
- Experiment Result
- Conclusion and Future Work
- References
- Query Recommendation by Modelling the Query-Flow Graph
- Introduction
- Related Work
- Query Recommendation
- Mixture Models
- Our Approach
- Query-Flow Graph
- Mixture Model on Query-Flow graph
- Intent-Biased Random Walk
- Experiments
- Data Set
- Evaluation of Intents
- Evaluation of Query Recommendation
- Conclusions
- References
- Ranking Content-Based Social Images Search Results with Social Tags
- Introduction
- Related Work
- Automatic Social Image Ranking
- Image-Tag Relationship Model
- Visual and Textual Descriptor
- Social Image Ranking
- Experiment
- Experimental Settings
- Parameter Selections
- Experimental Results
- Conclusion
- References
- User Study, Information Retrieval Evaluation and Interactive Information Retrieval
- Profiling a Non-medical Professional Searcher on a Medical Domain: What Do Search Patterns and Demographic Details Reveal?
- Introduction
- Related Work
- Research Methodology
- Pre-experiment Interview
- Simulated Work Task
- Observation
- Post- experiment Interview
- Demographic Details
- Results and Analysis
- Querying Behavior
- Search Results Evaluation Behaviour
- Querying versus Results Browsing Behaviour
- Discussion
- Future Work and Conclusion
- Reference
- Prioritized Aggregation of Multiple Context Dimensions in Mobile IR
- Introduction
- Background: Prioritized Multi-criteria Aggregation
- Problem Representation
- Prioritized Aggregation Operators
- Multidimensional Personalization of Mobile IR
- Topic
- Interest
- Location
- Experimental Evaluation
- Experimental Settings
- Results and Discussions
- Conclusion and Outlook
- References
- Searching for Islamic and Qur'anic Information on the Web: A Mixed-Methods Approach
- Introduction
- Related Work
- Patterns of Web Searching
- Islam and Web Searching
- Methodology
- Query Logs
- Interviews
- Results
- Query Log Analysis
- Interview data analysis
- Conclusion
- References
- Enriching Query Flow Graphs with Click Information
- Introduction
- Related Work
- The Model
- The Query Flow Graph
- Enriching the Query Flow Graph
- Query Recommendations
- Experimental Setup
- Search Logs
- Query Flow Graphs
- The Evaluation Framework
- Results and Discussion
- Conclusions
- Future Work
- References
- Effect of Explicit Roles on Collaborative Search in Travel Planning Task
- Introduction
- Experiment
- Travel Planning Task
- Independent Variable
- Dependent Variables
- Participants
- Procedure
- Results
- Travel Plans
- Search Behaviour
- Dialogue Development
- User Perceptions
- Conclusive Discussion
- References
- A Web 2.0 Approach for Organizing Search Results Using Wikipedia
- Introduction
- Search Results Organization
- Finding Wikipedia Page and Extracting Categories
- Expanding the Training Set and Gathering Search Results
- Experiments and Results
- Experiment Settings
- Conclusions and Future Works
- References
- Web Information Retrieval, Scalability and Adversarial Information Retrieval
- Recommend at Opportune Moments
- Introduction
- Related Work
- Problem Specification
- Motivation
- Group Dependency
- Main Idea
- Methodology
- The Proposed Approach
- Theory-Based Boundary $(t_2.5%, t_16%, t_50%, t_84%)$
- Global Optimal Boundary
- Local Optimal Boundary
- Experiment
- Dataset
- Evaluation Metric
- Experimental Result
- Discussion
- Conclusion
- Reference
- Emotion Tokens: Bridging the Gap among Multilingual Twitter Sentiment Analysis
- Introduction
- Related Work
- Study on Emotion Tokens in Tweets
- Types of Emotion Tokens
- Characteristics of Emotion Tokens in Twitter
- Multilingual Sentiment Lexicon Construction Based on Graph Propagation
- Co-occurrence Graph Construction of Emotion Tokens
- The Propagation and Smoothing Algorithm
- Sentiment Analysis with Emotion Tokens
- Experiments and Discussions
- Dataset
- Strategies of Comparative Evaluations
- Results and Discussions
- Conclusion and Future Work
- References
- Identifying Popular Search Goals behind Search Queries to Improve Web Search Ranking
- Introduction
- Observation and Main Idea
- Popular-Search-Goal-Based Search Model
- Search-Result Snippet Classification
- Search Goal Candidate Generation
- Popular Search Goal Validation
- Search-Goal-Based Ranking Model
- Performance Evaluation
- Experimental Setup
- Experimental Result
- Conclusion
- References
- A Novel Crawling Algorithm for Web Pages
- Introduction
- Background and Related Work
- FICA
- Proposed Algorithm
- FICA+
- Experimental Result
- Conclusion and Future Work
- References
- Extraction of Web Texts Using Content-Density Distribution
- Introduction
- Related Work
- Our Method
- Calculation of Word-Density Distribution in a Web Text
- Construction of Content-Density Distribution in a Web Text and Extraction of a Text
- Experiment
- Evaluating the Extracted Web Text
- Evaluating the Influence of the Extracted Web Text
- Conclusion
- References
- A New Approach to Search Result Clustering and Labeling
- Introduction
- An Approach to Search Result Clustering and Labeling
- Preprocessing
- Clustering
- Labeling via Term Weighting
- Performance Measures
- Clustering Evaluation
- Labeling Evaluation
- Experimental Results
- Conclusion
- References
- Efficient Top-k Document Retrieval Using a Term-Document Binary Matrix
- Introduction
- Related Work
- Preliminaries
- Model
- Combined Algorithm
- Our Approach
- The Key Idea
- Term-Document-Binary-Matrix-Based Combined Algorithm
- Experimental Evaluation
- Setup
- Results
- Conclusion
- References
- Machine Learning for Information Retrieval
- Topic Analysis for Online Reviews with an Author-Experience-Object-Topic Model
- Introduction
- Related Work
- Author-Experience-Object-Topic Model
- Motivation
- The Author-Experience-Object-Topic Model
- Gibbs Sampling Algorithms
- Generation of Variable x
- Variants of AEOT Model
- Experiments
- Dataset
- Experimental Setup
- Document Modeling
- Two Strategies about the Generation of x
- Discovered Topics
- Conclusion and Future Work
- References
- Predicting Query Performance Directly from Score Distributions
- Introduction
- Related Work
- Explicitly Modelling Query Performance
- Assumptions and Mixture Model
- Inferring Average Precision
- Estimating Parameters without Relevance Information
- Estimating Moments and Mixture
- Analysing Moments and Mixture
- Motivation and Improvement
- Expectation Maximisation Approach
- Results and Discussion
- Comparative Results
- Conclusion
- References
- Wikipedia-Based Smoothing for Enhancing Text Clustering
- Introduction
- Related Works
- Wikipedia-Based Smoothing
- Top-Feature Smoothing Algorithm
- Similarity Smoothing Algorithm
- Combination of the Two Methods
- Construction of the Document Similarity Graph and Link-Based Clustering
- Experiments and Results
- Evaluation Metrics
- Baselines
- K-Means Clustering of the Smoothed Documents
- Link-Based Clustering of the Smoothed Documents
- Experiments on 20-NG Dataset
- Conclusions and Future Works
- References
- ASVMFC: Adaptive Support Vector Machine Based Fuzzy Classifier
- Introduction
- Support Vector Machine
- Adaptive Support Vector Machine Based Fuzzy Classifier (ASVMFC)
- Clustering Phase
- SVM Training Phase
- Creating and Adjusting ASVMFC Phase
- Sample Reduction
- Experiments
- Conclusion
- References
- Ensemble Pruning for Text Categorization Based on Data Partitioning
- Introduction
- Related Work
- Experimental Environment
- Experimental Design
- Datasets
- Experimental Results
- Pruning Results
- Pruning-Related Parameters
- Conclusion
- References
- Sentiment Analysis for Online Reviews Using an Author-Review-Object Model
- Introduction
- Social Review Sentiment-Topic Model
- Motivation
- Model Formulation
- Parameter Estimation
- Sub Models of ARO
- Experiment
- Experimental Setup
- Define and Incorporate Sentiment Prior
- Sentiment Classification
- Topic Extraction
- Experiments for Chinese Reviews
- Conclusion and Future work
- References
- Natural Language Processing for Information Retrieval
- Semantic-Based Opinion Retrieval Using Predicate-Argument Structures and Subjective Adjectives
- Introduction
- Related Work
- Overview of Semantic-Based Opinion Retrieval
- Approaches Used for Semantic-Based Opinion Retrieval
- Grammatical Tree Derivations
- Predicate-Argument Structures
- Constructing a Semantic-Based Model for Opinion Retrieval
- Subjective Component Identification
- Semantic Similarity between Structures
- Transformed Terms Similarity (TTS)
- Linear Relevance Model (LRM)
- Experimental Results on Opinion Retrieval Task
- Conclusions and Future Work
- References
- An Aspect-Driven Random Walk Model for Topic-Focused Multi-document Summarization
- Introduction
- Related Works and Motivation
- Our Approaches
- Aspect-Based Model
- Random Walk Model
- Aspect-Driven Random Walk Model
- Evaluation Results and Analyses
- Task Description
- Data
- Automatic Evaluation
- Manual Evaluation
- Conclusion
- References
- An Effective Approach for Topic-Specific Opinion Summarization
- Introduction
- Representation of Topic-Specific Opinionated Information
- Pairwise Representation
- Weighting Scheme for Word Pair
- Word Pair Based TOS
- PageRank Based on Word Pair
- Summary Generation
- Evaluation
- Experiment Setting
- Performance Evaluation
- Related Work
- Conclusion and Future Work
- References
- A Model-Based EM Method for Topic Person Name Multi-polarization
- Introduction
- Related Work
- Method
- Model-Based Person Name Multi-polarization
- Off-Topic Block Elimination
- Weighted Correlation Coefficient
- Performance Evaluations
- Data Corpus and Evaluation Metric
- Effect of System Components
- Comparison with other Methods
- Multi-polarization Example
- Conclusions
- References
- Using Key Sentence to Improve Sentiment Classification
- Introduction
- Related Work
- Key Sentence Extraction
- Sentiment Attribute
- Position Attribute
- Special Words Attribute
- Classifier Combination with Key Sentences
- Co-training with Key Sentences
- Evaluation
- Experimental Setup
- Experimental Results
- Conclusion
- References
- Using Concept-Level Random Walk Model and Global Inference Algorithm for Answer Summarization
- Introduction
- Ranking Concepts Using Graph-Based Random Walk Model
- Graph-Based Random Walk Model on Text Content
- Two-Layer Link Graph for User Social Features
- Global Inference Algorithm for Answer Summarization
- Experiments and Discussion
- Experimental Setup
- Results
- Related Works
- Conclusion
- References
- Acquisition of Know-How Information from Web
- Introduction
- Related Work
- Acquisition of Know-How
- Extraction of Know-How Candidates
- Identification of Know-How
- Evaluation
- Preparation
- Settings
- Experimental Results
- Discussion
- Open-Domain Tests
- Conclusion
- References
- Topic Based Creation of a Persian-English Comparable Corpus
- Introduction
- Constructing the Comparable Corpus
- Base Features Selection and Construction of the MI Graph
- Extraction of Major Topics Using MI Graph
- Construction of Document Clusters
- Extraction of Related Topics in the Persian Collection
- Document Alignment
- Experiments and Results
- Creating and Evaluating the Comparable Corpus
- Extracting Word Associations
- Cross-Language Information Retrieval Experiments
- Conclusions and Future Work
- References
- A Web Knowledge Based Approach for Complex Question Answering
- Introduction
- Complex Question Answering Based on Web Knowledge Bases
- Summarization from Web Knowledge Bases
- Answer Sentence Acquisition Using Topic Model
- Answer Ranking
- Experimental Results
- Conclusion
- References
- Learning to Extract Coherent Keyphrases from Online News
- Introduction
- Related Work
- Learning to Extract Keyphrases
- Learning to Rank
- Two-Phase Ranking Approach to Keyphrase Extraction
- Features in Two Phases
- Experiment
- Experimental Data
- Evaluation Measures
- Performance of Keyphrase Extraction
- Comparing with other Algorithms
- Discussion
- Conclusion
- References
- Maintaining Passage Retrieval Information Need Using Analogical Reasoning in a Question Answering Task
- Introduction
- Question Answering Approach
- Bayesian Analogical Reasoning
- Experimental Setting
- Results and Discussions
- Performance of Query Expansion
- Indri Pseudo-relevance Feedback
- Question Type and Retrieval Performance Issues
- Conclusions and Future Work
- References
- Improving Document Summarization by Incorporating Social Contextual Information
- Introduction
- The Proposed Approach
- Social Context Recognition
- Sentence Ranking Based on Social Context
- Sentence Extraction
- Experiments
- Data Set
- Evaluation Methods
- Experimental Results
- Impact of the Adjusting Parameter $?$
- Conclusion and Future Work
- References
- Automatic Classification of Link Polarity in Blog Entries
- Introduction
- Related Work
- Sentiment Analysis
- Lexicon for Sentiment Analysis
- Classification of Link Polarity
- Classification of Link Polarity
- Link Polarity
- Classification of Link Polarity
- Experiments
- Data Sets and Experimental Setting
- Evaluation of the Link Lexicon
- Evaluation of Citing Areas
- Conclusion
- References
- Feasibility Study for Procedural Knowledge Extraction in Biomedical Documents
- Introduction
- Related Work
- Methodology for Procedural Knowledge Extraction
- Modeling
- Extraction Procedures
- Target Documents
- Training Corpus
- Experiments
- Preprocessing
- Purpose/Solution Sentence Classification
- TAM Identification
- TAM Association
- Relation Extraction
- Results
- Results on Purpose/Solution Sentence Classification
- Results on TAM Identification
- Results on TAM Association and Relation Extraction
- Conclusion
- References
- Arabic Script Text Processing and Retrieval
- Small-Word Pronunciation Modeling for Arabic Speech Recognition: A Data-Driven Approach
- Introduction
- Motivation
- The Baseline System
- Arabic Phoneme Set
- Phonetic Dictionary
- The Proposed Method
- Testing and Evaluation
- Conclusion
- References
- The SALAH Project: Segmentation and Linguistic Analysis of $?adi?$ Arabic Texts
- Introduction
- Contents and Structure of the Corpus: The $?adi?s$
- Extracting Surface's Information: From Segmentation to Representation
- $?adi?'s$ Segmentation: Pairing Explicit and Implicit Information
- Extraction and Organization: The HadExtractor Program
- Representation: Transmitters' Chains and Graphs
- Analyzing the Corpus: The Revised AraMorph Analyzer
- The Original AraMorph Implementation
- Modifications to the Algorithm
- Results and Evaluation
- Further Research
- References
- Exploring Clustering for Multi-document Arabic Summarisation
- Introduction
- Related Work
- Multi-document Summarisation
- Clustering for Summarisation
- Test Collections
- Evaluation
- Arabic Translation of the DUC-2002 Dataset
- Cluster-Based Summarisation
- K-means Clustering
- Experiment 1: Clustering All Sentences
- Experiment 2: Clustering for Redundancy Elimination
- Experimental Setup
- Results and Discussion
- Conclusion
- References
- ZamAn and Raqm: Extracting Temporal and Numerical Expressions in Arabic
- Introduction
- General Background
- Temporal Expressions (TMPs)
- Numerical Expressions (NUMs)
- Evaluation of Temporal and Numerical Expressions
- Related Work
- ZamAn: Temporal Expressions Labeller
- Preprocessing the ATB
- Classification Features
- ZamAn: Experimental Results
- Raqm: Numerical Expressions Labeller
- Normalisation of NUMs
- Raqm: Experimental Results
- Conclusion
- References
- Extracting Parallel Paragraphs and Sentences from English-Persian Translated Documents
- Introduction
- Related Works
- Extracting Parallel Paragraphs and Sentences
- Feature Similarities
- Combining Similarities
- Using Genetic Algorithm for Weight Learning
- Experiments and Results
- Experiment 1: Paragraph Alignment
- Experiment 2: Sentence Alignment
- Conclusion
- References
- Effect of ISRI Stemming on Similarity Measure for Arabic Document Clustering
- Introduction
- Related Work
- Methodology
- Arabic Text Pre-processing
- Arabic Stemming Algorithm
- Term Representation
- Similarity Measures
- Evaluation
- Clustering
- Data Description and Performance Measure
- Experiments and Result
- Conclusion and Future Work
- References
- A Semi-supervised Approach for Key-Synset Extraction to Be Used in Word Sense Disambiguation
- Introduction
- Related Work
- Word Sense Disambiguation Approach
- Persian Word Sense Disambiguation
- Key-Synset Extraction Approach
- Experimental Results
- Training Phase
- Building the Test Corpus
- Evaluation of the Method
- Conclusion and Future Works
- References
- Mapping FarsNet to Suggested Upper Merged Ontology
- Introduction
- Resources
- Princeton WordNet
- FarsNet
- SUMO
- Our Mapping Methodology
- Adjectives
- Nouns
- Verbs
- Discussion
- Mapping Output
- Mapping Rules
- References
- Topic Detection and Multi-word Terms Extraction for Arabic Unvowelized Documents
- Introduction
- Related Works
- Topic Detection System
- Documents Pre-treatment
- Vocabulary Oriented Topic Generation
- Topic Detection
- Multi-word Terms Extraction System
- Linguistic Filter
- Statistical Filter
- Experimentation and Results
- Topic Detection
- MWTs Extraction
- Conclusion
- References
- Author Index
