Advances in Information Retrieval Theory

Name: Advances in Information Retrieval Theory | Third International Conference, ICTIR 2011, Bertinoro, Italy, September 12-14, 2011, Proceedings
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

Third International Conference, ICTIR 2011, Bertinoro, Italy, September 12-14, 2011, Proceedings

Giambattista Amati Fabio Crestani(Herausgeber*in)

Springer (Verlag)

Erschienen am 8. September 2011

XVI, 346 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-642-23318-0 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Title
Preface
Organization
Table of Contents
Invited Talks
Axiomatic Analysis and Optimization of Information Retrieval Models
What Is Quantum Information Retrieval?
Predicting Query Performance
User Perspectives on Query Difficulty
Introduction
Related Work
Crowdsourcing User Perspectives
Query Difficulty Estimation as Opinion Fusion
Turning User Perspectives into Formal Opinions
Fusing Opinions of Query Difficulty Using Bayesian Consensus
Bayesian Consensus Assessments versus TREC Categories
Conclusion
References
A Unified Framework for Post-Retrieval Query-Performance Prediction
Introduction
Related Work
Query-Performance Prediction Framework
Deriving Previously Proposed Predictors
Experiments
Experimental Setup
Experimental Results
Summary and Future Work
References
Predicting the Performance of Recommender Systems: An Information Theoretic Approach
Introduction
Performance Prediction in Information Retrieval
Predictive Models of Recommendation Performance
Ground Models
Qualitative Observation
Experiments
Conclusion
References
Latent Semantic Analysis and Word Co-occurrence Analysis
Trading Spaces: On the Lore and Limitations of Latent Semantic Analysis
Introduction
How `Semantic' Is LSA?
US Patent No. 4,839,853
Semantics and Cross-Language Retrieval
The Notion of `Lexical Noise' in LSA
SVD for LSA?
US Patent No. 6,687,696, pLSI, and LDA
Locating Semantic Factors
Parsimonious Alternatives to LSA
How l0 and l1 Norms Are Better Than LSA
Conclusion
References
Quantum Latent Semantic Analysis
Introduction
Quantum Information Retrieval
Quantum Latent Semantic Analysis
QLSA vs. LSA
QLSA vs. PLSA
Experimental Evaluation
Collections
Dimensions of the Latent Space
Recall-Precision Evaluation
Discussion and Conclusions
References
Pure High-Order Word Dependence Mining via Information Geometry
Introduction
Related Work
Preliminaries of Information Geometry
Coordinates of Probability Distributions
Coordinate Orthogonality
Coordinate Parameter Estimation
The Spectrum of High-Order Pure Dependence
Implementation and Complexity Analysis
Application
An Extended Relevance Model
Experimental Setup
Results and Analysis
Conclusions and Future Work
References
Query Expansion and Re-ranking
Promoting Divergent Terms in the Estimation of Relevance Models
Introduction and Motivation
Background
Language Modelling for the Initial Ranking
Relevance Models
Kullback Leibler Divergence for Pseudo Relevance Feedback
Promoting the Divergence in Pseudo Relevance Feedback
Kullback Leibler Divergence Based Query Expansion in the Language Modelling Framework
Relevance Models with Promotion of Divergent Terms
Experiments and Results
Collections
Methods
Training and Evaluation
Results
Related Work
Conclusions and Future Work
References
Is Document Frequency Important for PRF?
Introduction
Some Statistics on PRF
The Document Frequency Constraint
Validation of the DF Constraint
Review of PRF Models
Well-Founded, Simple PRF Reweighting
Related Work
Conclusion
References
Comparison of Information Retrieval Systems and Approximate Search
Model-Based Inference about IR Systems
Introduction
Traditional Model-Based Inference
Model-Based Evaluation Measures
Statistical Hypothesis Tests
Bayesian Inference
Bayesian Linear Model
Direct Inference about Relevance
Modeling Other Evidence
Empirical Analysis
Experimental Set-Up
Classical T-Test vs. Bayesian Linear Model vs. Direct Inference
Advanced Analysis
Conclusion
References
Selecting a Subset of Queries for Acquisition of Further Relevance Judgements
Introduction
Related Work
Expanding Relevance Judgements
The Query Selection Problem
Experimental Evaluation
Generalization
Comparing Two Relevance Judgment Allocation Methods
Discussion and Conclusion
References
On the Feasibility of Unstructured Peer-to-Peer Information Retrieval
Introduction
Probabilistic Search
Ferreira et al's Model
Cohen and Shenkers' Model
Probably Approximately Correct Search
Non-uniform Replication in PAC
Communication Cost
Communications Load for Uniform Replication
Communication Load for Non-uniform Replications
Conclusion
References
Probability Ranking Principle and Alternatives
Can Information Retrieval Systems Be Improved Using Quantum Probability?
Introduction
Related Work
Quantum Probability and Decision
Optimal Projectors in the Quantum Space
Quantum Probability Gives a Different Ranking
Future Developments and Conclusions
References
An Analysis of Ranking Principles and Retrieval Strategies
Introduction
Principles and Strategies
Alternatives to the PRP
Parametric Instantiations of iPRP and qPRP
Analysis of Ranking Behaviours
Kinematics of Documents
Summary and Future Work
References
Towards a Better Understanding of the Relationship between Probabilistic Models in IR
Introduction
Notation and Definitions
Gap1 - Between the PRP and PR Models
The PRP
PR Models
A Bridge for Gap1
Gap2 - Between PR Models and Language Models
Language Models
Differences between PR Models and Language Models
Bridging Gap2 at the Functional and Ranking Level
Conclusions
References
Interdisciplinary Approaches
Cognitive Processes in Query Generation
Introduction
Related Work
Methodology
Experiments
Conclusions
References
Protocol-Driven Searches for Medical and Health-Sciences Systematic Reviews
Introduction
Systematic Reviews
Issues with the Reporting of Clinical Outcomes
IR Searches in Systematic Reviews
Protocols for Systematic Reviews
Systematic Review Case Study
Motivation and Methodology
Results Overview
Towards Protocol-Driven IR
Recall-Focused Search Tasks
Roadmap of IR Research for Systematic Reviews
Evaluation Methodologies
Conclusions
References
Enhanced Information Retrieval Using Domain-Specific Recommender Models
Introduction
Related Work
Personalized Information Retrieval
Recommender System
Combining Information Retrieval with Domain-Specific Recommender Models
Experiment
Data Collection
Experiment Setup
Result
Conclusion and Future Work
References
User and Relevance
Exploring Ant Colony Optimisation for Adaptive Interactive Search
Introduction
Related Work
ACO for Learning from Query Logs
The Evaluation Framework
The Experimental Setup
Search Log Data
Experiments
Results
Discussion and Future Work
References
Applying the User-over-Ranking Hypothesis to Query Formulation
Introduction
Use Cases
Related Work
Notation and Basic Definitions
Baseline Strategies for Promising Queries
Computing All Promising Queries
Co-occurrence-Informed Baseline
Heuristic Search Strategies
Experimental Analysis
Known Item Finding
Search Session Support
Empty Result Lists
Conclusion and Outlook
References
How to Count Thumb-Ups and Thumb-Downs: User-Rating Based Ranking of Items from an Axiomatic Perspective
Introduction
Problem
Popular Methods
Difference
Proportion
Wilson Interval
Proposed Approach
Additive Smoothing
Interpolation Smoothing
Other Smoothing Techniques
Generalisations
Axiomatic Examination
Conclusions
References
Result Diversification and Query Disambiguation
Aggregated Search Result Diversification
Introduction
Related Work
Search Result Diversification
Aggregated Search
The Nature of Ambiguity
Problem Formulation
Modelling Aggregated Diversification
Evaluating Aggregated Diversification
Conclusions
References
Topical Categorization of Search Results Based on a Domain Ontology
Introduction
Related Work
Definition of a Topical Granular Taxonomy from a Domain Ontology
The Proposed Method
The Categorization Process
Preliminary Experiments
Conclusions and Future Work
References
Towards Semantic Category Verification with Arbitrary Precision
Introduction
Prior Work
Framework
Seed Identification
Training Using the Seeds
Answering the Question
Empirical Evaluation
Data Sets
Results and Discussion
Conclusions
References
Logical Operators and Descriptive Approaches
Negation for Document Re-ranking in Ad-hoc Retrieval
Introduction
A Re-ranking Method Based on Non-relevant Documents
Semantic Document Space
Re-ranking Using Quantum Negation
Evaluation
Related Work
Conclusions
References
A Descriptive Approach to Classification
Introduction and Motivation
Background and Related work
Standard Classifiers
Descriptive Approaches
Probabilistic Datalog
Modelling Classification in Probabilistic Datalog
Proving the Correctness of PD Programs
Feasibility Study
Experiment Set-Up
Results
Discussion and Future Work
References
Posters
Do Subtopic Judgments Reflect Diversity?
Introduction
Method
Results
Conclusions and Future Directions
References
On Upper Bounds for Dynamic Pruning
Introduction
Experimental Setup
Experimental Results
Conclusions
References
A Comparative Study of Pseudo Relevance Feedback for Ad-hoc Retrieval
Introduction
Related PRF Methods
Experiments
Conclusions and Future Work
References
A Generic Data Model for Schema-Driven Design in Information Retrieval Applications
Introduction and Motivation
The Data Model
Modelling Layers
Probability Estimation
Conclusion
References
A Query-Basis Approach to Parametrizing Novelty-Biased Cumulative Gain
Introduction
Analysis of -NDCG
Deriving a Threshold for
Remarks and Conclusion
References
Investigating Query-Drift Problem from a Novel Perspective of Photon Polarization
Introduction
Quantum-Inspired Approach
Photon Polarization
QM-Inspired Fusion Approach
Empirical Evaluation
Conclusion
References
Using Emotion to Diversify Document Rankings
Introduction
Approach
Diversifying Document Rankings
Construction of Emotion Vectors
Experiment and Results
Conclusions
References
Improved Stable Retrieval in Noisy Collections
Introduction
Methodology
Conclusions
References
On the use of Complex Numbers in Quantum Models for Information Retrieval
Introduction
Use of Complex Numbers in Quantum Theory
Analysis of the Potentials of Complex Numbers for IR
Conclusions
References
A Query Performance Analysis for Result Diversification
Introduction
Method
Experiments and Results
Discussion and Conclusion
References
Rare Disease Diagnosis as an Information Retrieval Task
Introduction
Retrieving Rare Diseases: Preliminary Study
The Characteristics of Rare Disease Retrieval
Conclusion
References
Distilling Relevant Documents by Means of Dynamic Quantum Clustering
Introduction
Dynamic Quantum Clustering
Experiments and Results
References
Adding Emotions to Pictures
Introduction
Children: Colors and Emotions
Adding Emotions to Simple Pictures
Conclusions
References
Author Index

Inhalt (PDF)

Systemvoraussetzungen

Als PDF speichern Als Link merken