Advances on Data Mining: Applications and Theoretical Aspects
11th Industrial Conference, ICDM 2011, New York, NY, USA, August 30 - September 3, 2011, Proceedings
Published on 9. August 2011
XI, 329 pages
978-3-642-23184-1 (ISBN)
Description
This book constitutes the refereed proceedings of the 11th Industrial Conference on Data Mining, ICDM 2011, held in New York, USA in September 2011.
The 22 revised full papers presented were carefully reviewed and selected from 100 submissions. The papers are organized in topical sections on data mining in medicine and agriculture, data mining in marketing, data mining for Industrial processes and in telecommunication, Multimedia Data Mining, theoretical aspects of data mining, Data Warehousing, WebMining and Information Mining.
More details
Other editions
Additional editions
Petra Perner
Advances on Data Mining: Applications and Theoretical Aspects
11th Industrial Conference, ICDM 2011, New York, NY, USA, August 30 - September 3, 2011, Proceedings
Book
08/2011
1st Edition
Springer
€53.49
Shipment within 7-9 days
Content
- Title
- Preface
- Organization
- Table of Contents
- Theoretical Aspects of Data Mining
- Improvements over Adaptive Local Hyperplane to Achieve Better Classification
- Introduction
- Adaptive Local Hyperplane and Feature Weighting Scheme
- A Numerical Correction
- Experimental Results
- Conclusion
- References
- Prognostic Models Based on Linear Separability
- Introduction
- Linear Regression Models and Learning Sets with Different Structure
- Learning Sets in Survival Analysis
- Linear Separability of Two Data Sets
- CPL Penalty and Criterion Functions for Interval and Ranked Regression
- Relaxed Linear Separability (RLS) Method of Feature Selection for Prognostic Models
- Concluding Remarks
- References
- One Class Classification for Anomaly Detection: Support Vector Data Description Revisited
- Introduction
- Support Vector Data Description Revisited
- Outlier Detection as an Optimization Problem
- Properties of the Solution
- Linear Slacks and Linear Loss
- Basic Analysis
- Further Properties
- Conclusions
- References
- How to Interpret Decision Trees?
- Introduction
- The Problem
- Decision Tree Induction Based on the Gain Ratio (Entropy-Based Measure)
- Attribute Splitting Criteria
- How to Interpret the Results of a Decision Tree
- Quantitative Measures of the Quality of the Decision Tree Model
- Explanation Capability of the Decision Tree Model
- Revision of the Data Label
- Comparison of Two Decision Trees
- Conclusions
- References
- Comparing Classifiers and Metaclassifiers
- Introduction
- Related Work
- Ensemble Methods
- Combination of Generative and Non-generative Ensemble
- Experimental Results
- Conclusions
- References
- Fast Data Acquisition in Cost-Sensitive Learning
- Introduction
- Related Work
- The Framework of Data Acquisition
- Fast Data Acquisition Strategy (FDA)
- Experiments
- Conclusions and Future Work
- References
- Data Mining in Medicine
- Application of a Unified Medical Data Miner (UMDM) for Prediction, Classification, Interpretation and Visualization on Medical Datasets: The Diabetes Dataset Case
- Introduction
- Overview of Data Mining Algorithms Used in the Medical Data Miner
- Neural Networks
- Decision Tree (C4.5) Algorithm
- k-Means Clustering Algorithm
- Data Visualization
- Methodology
- Results and Discussion
- Conclusion
- References
- Melanoma Diagnosis and Classification Web Center System: The Non-invasive Diagnosis Support Subsystem
- Introduction
- Structure and Operation of the System
- Recognition Algorithms and Classification Process
- Learning Model Based on a Classic and Optimized ABCD Rule
- Learning Model in Form of Decision Tree
- Learning Model Based on the Genetic Dichotomization
- Learning Model in Form of Belief Network
- Algorithm for Optimal Diagnosis Selection
- Conclusions and Future Remarks
- References
- Characterizing Cell Types through Differentially Expressed Gene Clusters Using a Model-Based Approach
- Introduction
- Material
- Methods
- Model-Based Clustering
- Gibbs Sampler Algorithm for Cluster Optimization
- Results
- Cluster Stability and Quality
- Prediction Accuracy
- Marker Module Detection and Analysis
- Summary
- References
- Experiments with Hybridization and Optimization of the Rules Knowledge Base for Classification of MMPI Profiles
- Introduction
- MMPI Profiles
- The Copernicus System
- The Rules Knowledge Base Hybridization
- The Rules Knowledge Base Optimization
- Experiments
- Conclusions
- References
- Multimedia Data Mining
- Unsupervised Classification of Hyperspectral Images on Spherical Manifolds
- Introduction
- Spherical Embedding of Image Pixels
- von Mises-Fisher Model and Bayes Rule
- Maximum Likelihood Estimation
- Mixture of von Mises-Fisher Model
- Experiments
- Data
- Results
- Conclusions
- References
- Recognition of Porosity in Wood Microscopic Anatomical Images
- Introduction
- Materials and Methods
- Image Data
- Segmentation Algorithm
- Feature Extraction
- Porosity Recognition
- Results and Discussion
- Classification Performance
- Discussion
- Conclusion
- References
- Data Mining in Agriculture
- Exploratory Hierarchical Clustering for Management Zone Delineation in Precision Agriculture
- Introduction
- Article Structure
- Data Set and Existing Literature on Spatial Clustering
- Precision Agriculture Data Description
- Review of Existing Spatial Clustering Algorithms
- Hierarchical Clustering with Spatial Constraints
- Phase 1: Spatial Field Tessellation via k-Means (Optional)
- Phase 2: Merging Clusters
- Experimental Setup and Results
- Limitations and Parameter Guidelines
- Conclusion and Discussion
- Future Work
- References
- High Classification Rates for Continuous Cow Activity Recognition Using Low-Cost GPS Positioning Sensors and Standard Machine Learning Techniques
- Introduction
- Related Work
- Collection of Position Data for Cow Activities
- Recognition of Cow Activities
- Results
- Discussion
- Conclusion
- References
- Mining Pixel Evolutions in Satellite Image Time Series for Agricultural Monitoring
- Introduction
- Grouped Frequent Sequential Patterns
- Preliminary Definitions
- Sequential Patterns
- Spatial Connectivity
- Grouped Frequent Sequential Pattern Extraction
- Experiments
- The ADAM SITS: Presentation, Selection and Preprocessing
- Quantitative Results
- Qualitative Results
- Related Work
- Conclusion
- References
- Data Mining for Industrial Processes
- Robust, Non-Redundant Feature Selection for Yield Analysis in Semiconductor Manufacturing
- Introduction
- History of Analysis Approach
- Feature Selection
- Analysis
- Discussion of Results
- Conclusions
- References
- Integrated Use of ICA and ANN to Recognize the Mixture Control Chart Patterns in a Process
- Introduction
- The Methodologies
- Independent Component Analysis
- Artificial Neural Network
- The Proposed Approach and the Example
- The Proposed ICA-ANN Scheme
- Simulated Experiments
- Conclusion
- References
- Optimized Fuzzy Decision Tree Data Mining for Engineering Applications
- Introduction
- Literature Review
- Proposed Inference Mechanism
- Numerical Example
- Conclusion
- References
- Data Warehousing
- Graph-Based Data Warehousing Using the Core-Facets Model
- Introduction
- Related Work
- Core-Facets Model
- Data Gathering
- Data Interpretation
- Data Preparation
- Implementation and Application of Model
- Ontology
- Gathering and Interpreting Sensor Data (Phases 1 and 2)
- Creating Facets (Phase 3)
- Using Facets in Analysis
- Summary
- References
- Data Mining in Marketing
- General Sales Forecast Models for Automobile Markets Based on Time Series Analysis and Data Mining Techniques
- Introduction
- Data and Workflow
- Methodology
- Calendar Component Estimation
- Seasonal Component Estimation
- Trend Component Estimation
- Results
- Comparison of Data Mining Methods
- Absolute and Relative Exogenous Parameters
- Explicability of the Results
- Application to the US-American Automobile Market
- Conclusions
- References
- WebMining/Information Mining
- Towards a Spatial Instance Learning Method for Deep Web Pages
- Introduction
- Related Work
- Positional Document Object Model
- Preliminary Definitions
- PDOM Definition
- PDOM Building
- Visual Similarity
- Instance Learning Algorithm
- Data Region and Data Record Identification
- Data Records and Data Item Extraction
- Empirical Evaluation
- Conclusions and Future Work
- References
- Data Mining in Telecommunications
- Applying User Signatures on Fraud Detection in Telecommunications Networks
- Introduction
- Fraud in Telecommunications Networks
- Fraud Losses
- Possible Fraud Cases
- User Profiling Based on Signatures
- Profiling
- About Signatures
- Signatures in Telecommunications Fraud
- Distance Functions
- Measures and Weights
- System's Architecture and Prototype Integration
- Summaries Processing
- Experiments and Results
- Conclusions and Future Work
- References
- Aspects of Data Mining
- Methods in Case-Based Classification in Bioinformatics: Lessons Learned
- Introduction
- Methods
- Classification Algorithms
- Feature Selection Algorithms
- Evaluation Methods
- Results
- Datasets
- Software and Hardware
- Algorithms
- Performance on All Features
- Performance on Selected Features
- Summary of Results and Discussion
- Related Works
- Conclusion
- References
- Prediction of Batch-End Quality for an Industrial Polymerization Process
- Introduction
- Industrial Installation
- Data Preprocessing
- Data Alignment Procedure
- Data Alignment Results
- Model Identification
- Partial Least Squares Modelling
- Optimal Input Selection
- Model Order Determination
- Model Training
- Intermediate Model Identification
- Validation Results
- Conclusions
- References
- Author Index
