Inductive Logic Programming

Name: Inductive Logic Programming | 20th International Conference, ILP 2010, Florence, Italy, June 27-30, 2010, Revised Papers
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

20th International Conference, ILP 2010, Florence, Italy, June 27-30, 2010, Revised Papers

Paolo Frasconi Francesca A. Lisi(Herausgeber*in)

Springer (Verlag)

Erschienen am 13. Juni 2011

XI, 278 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-642-21295-6 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

6489
Preface
Organization
Table of Contents
Abstracts of Invited Talks
Rule Interchange Format : Logic Programming's Second Wind?
Practical Probabilistic Programming
Introduction
References
Probabilistic Relational Learning and Inductive Logic Programming at a Global Scale
References
Research Papers
Learning Multi-class Theories in ILP
Introduction
Motivation
Improved Learning of Multi-class Theories
Empirical Evaluation
Related Work
Concluding Remarks
References
A Numerical Refinement Operator Based on Multi-Instance Learning
Introduction and Background
Method
Experiments
Datasets
Experimental Setup
Experimental Results
Conclusion
References
Not Far Away from Home: A Relational Distance-Based Approach to Understanding Images of Houses
Introduction
Setting
Approach
Experimental Setup and Results
Conclusions
References
Approximate Inference for Logic Programs with Annotated Disjunctions
Introduction
Logic Programs with Annotated Disjunctions
Approximate Inference
k-best Algorithm
Monte Carlo Algorithm
Experiments
References
Approximate Bayesian Computation for the Parameters of PRISM Programs
Introduction
Approximate Bayesian Computation
PRISM
ABC for PRISM
Choice of Prior Distribution
Choice of Distance Function
Choice of Perturbation Kernel
Experimental Results
Conclusions and Future Work
References
Probabilistic Rule Learning
Introduction
Problem Specification
Analysis
ProbFOIL: A Probabilistic First Order Rule Learner
Experiments
Surfing
Eulisis
Conclusions
References
Interactive Discriminative Mining of Chemical Fragments
Introduction
Background
The iLogCHEM System
Integrating Structural Information in the Search
Macros
Molecular Properties
Interactive Search and Refinement
Conclusions
References
MMRF for Proteome Annotation Applied to Human Protein Disease Prediction
Introduction
Modular Multi-Relational Framework
MMRF Applied to Human Protein Disease Prediction
Results and Discussion
Conclusions and Further Work
References
Extending ProbLog with Continuous Distributions
Introduction
ProbLog
Hybrid ProbLog
Distribution Over Continuous Subprograms
Success Probabilities of Queries
Exact Inference
Experiments
Conclusions and Future Work
References
Multivariate Prediction for Learning on the Semantic Web
Introduction
Related Work
Statistical Modeling
Defining the Sample
The Random Variables in the Data Matrix
Non-random Covariates in the Data Matrix
Algorithms for Learning with Statistical Units Node Sets
Experiments
Data Set and Experimental Setup
Results
Conclusions and Outlook
References
Learning Action Descriptions of Opponent Behaviour in the Robocup 2D Simulation Environment
Introduction
Action Descriptions for Dynamic Domains
Efficient Induction of Action Laws with IAction
Modeling Opponent Behaviour
Discretization of the Environment
Representation
Learning Results
Previous Work and Conclusions
References
Hypothesizing about Causal Networks with Positive and Negative Effects by Meta-level Abduction
Introduction
Meta-level Abduction
Reasoning about Positive and Negative Causal Effects
Alternating Axioms for Causality
Axiomatization with Default Assumptions
Case Study: p53 Signal Networks
Enumerating Tumor Suppressors
Recovering Links in CDK Networks
Discussion and Related Work
Conclusion
References
BET : An Inductive Logic Programming Workbench
Introduction
Overview
Design of BET
Need for Standard Format
BET Standard Format
Positive Example File:
Negative Example File:
Background knowledge File:
Language Restriction File:
Class Diagram and Components of BET
Support for Building ILP Systems in BET
Integrating a Legacy System in BET
Implementing a New ILP System in BET
Related Work
Aleph
GILPS
Advantages of BET over Existing Systems
Summing Up BET
References
Seeing the World through Homomorphism: An Experimental Study on Reducibility of Examples
Introduction
Preliminaries
Reducing the Examples
Experiments
Mutagenesis
Predictive Toxicology Challenge
CAD
Aleph with Reduced and Non-reduced Data
Conclusions
References
Learning Discriminant Rules as a Minimal Saturation Search
Introduction
Preliminaries, Notations and Saturation
Saturation of an Example
Minimal Saturation to Search
Refinement Operator
Refinements Selection
Learning a Rule
Advantages and Limitations
Application to Constraint Satisfaction Problem and Experiments
Conclusion
References
Variation of Background Knowledge in an Industrial Application of ILP
Introduction
Application Descriptions
University Innovation Centre (UIC) Overview
Tomato Application
Predictive Toxicology
Ondex: A Biological Background Knowledge Generator
Experiments
Materials and Methods
Results and Discussion
Conclusions and Further Work
References
Pruning Search Space for Weighted First Order Horn Clause Satisfiability
Introduction
Satisfiability in Horn Clauses
T Operator
Modified_T Step
Modified_Weighted_SAT
Results
Conclusion and Future Work
References
Multi-relational Pattern Mining Based-on Combination of Properties with Preserving Their Structure in Examples
Introduction
Patterns and Mapix Algorithm
Ideas and an Algorithm
Experiments and Concluding Remarks
References
Learning from Noisy Data Using a Non-covering ILP Algorithm
Introduction
Related Work
HYPER/N
Handling of Noisy Data in HYPER/N
Learning Multiple Clauses from Noisy Data
Some Additional Improvements
Experiments with Weather Data
Conclusion
References
Can HOLL Outperform FOLL?
Introduction and Motivations
?Progol: A Higher-Order ILP System
Results and Applications
Future Works
References
Incremental Learning of Relational Action Models in Noisy Environments
Introduction
Learning Problem
Action Model
Relational Representation
Matching and Covering
Incremental Relational Learning of an Action Model
Sketch of the Algorithm
Conservative Generalizations and Specializations
Covering and Elimination of Irrelevant Rules
Prediction with the Model
Empirical Study
Conclusion
References
When Does It Pay Off to Use Sophisticated Entailment Engines in ILP?
Introduction and Motivation
The -Subsumption Problem
Subsumption versus Resolution
Entailment Algorithms in ProGolem
Time Complexity
Empirical Evaluation
Materials and Methods
Results and Discussion
Conclusions and Future Directions
References
Stochastic Refinement
Introduction
Preliminaries
Stochastic Refinement Operators
Stochastic Refinement Search
Stochastic Refinement Search as a Gibbs Sampling Algorithm
Stochastic Refinement Search as a Random Heuristic Search
Examples of Stochastic Refinement Search
Stochastic Refinement Relative to a Bottom Clause
Divisibility of the Subsumption Lattice Relative to ?
Conclusions
References
Fire! Firing Inductive Rules from Economic Geography for Fire Risk Detection
Introduction
Related Work
The Fire DataSet
Methodology
The Background Knowledge
Results and Discussion
Conclusions
References
Automating the ILP Setup Task : Converting User Advice about Specific Examples into General Background Knowledge
Introduction
The Onion
Converting Advice to Background Knowledge
Experiments
Test Beds
Results and Discussion
Related Work
Conclusions and Future Work
References
Speeding Up Planning through Minimal Generalizations of Partially Ordered Plans
Introduction
Method
Experiments
Conclusions
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken