New Frontiers in Mining Complex Patterns

Name: New Frontiers in Mining Complex Patterns | Third International Workshop, NFMCP 2014, Held in Conjunction with ECML-PKDD 2014, Nancy, France, September 19, 2014, Revised Selected Papers
Brand: Springer
Price: 50.28 EUR
Availability: OnlineOnly

Third International Workshop, NFMCP 2014, Held in Conjunction with ECML-PKDD 2014, Nancy, France, September 19, 2014, Revised Selected Papers

Annalisa Appice Michelangelo Ceci Corrado Loglisci Giuseppe Manco Elio Masciari Zbigniew W. Ras(Editor)

Springer (Publisher)

Published on 27. April 2015

XII, 211 pages

E-Book

PDF with digital watermarking

System requirements

978-3-319-17876-9 (ISBN)

€50.28incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Persons

Content

Intro
Preface
Organization
Sampling and Presenting Patterns from Structured Data
Contents
Classification and Regression
Semi-supervised Learning for Multi-target Regression
1 Introduction
2 MTR with Ensembles of Predictive Clustering Trees
2.1 Predictive Clustering Trees for MTR
2.2 Ensembles of PCTs
3 Self-training for MTR with Ensembles of PCTs
4 Experimental Design
4.1 Data Description
4.2 Experimental Setup and Evaluation Procedure
5 Results and Discussion
6 Conclusions
References
Evaluation of Different Data-Derived Label Hierarchies in Multi-label Classification
1 Introduction
2 Background
2.1 The Task of Multi-label Classification (MLC)
2.2 The Task of Hierarchical Multi-label Classification (HMC)
3 The Use of Data Derived Label Hierarchies in Multi-Label Classification
3.1 Generating a Label Hierarchy on a Multi-label Output Space
3.2 Solving MLC Problems by Using Classification Approaches for HMC
3.3 Classification Approaches for HMC
4 Experimental Design
4.1 Datasets and Evaluation Measures
4.2 Experimental Setup
4.3 Statistical Evaluation
5 Results and Discussion
6 Conclusions and Further Work
A Evaluation Measures
A.1 Example Based Measures
A.2 Label Based Measures
A.3 Ranking Based Measures
B Complete Results from the Experimental Evaluation
References
Clustering
Predicting Negative Side Effects of Surgeries Through Clustering
1 Introduction
2 Background
2.1 Multi-valued Information System
2.2 Atomic Action Terms and Action Terms
2.3 Meta-Actions for Multi-valued Information System
3 Negative Side Effects
4 Clustering Based on Negative Side Effects
5 New Approach for Predicting Negative Side Effects
5.1 Distance Between Two Patients
5.2 Distance Between a Patient and a Cluster
6 Dataset and Experiments
6.1 HCUP Dataset Description
6.2 Experiments
7 Summary and Conclusions
References
Parallel Multicut Segmentation via Dual Decomposition
1 Introduction
2 Related Work
3 Segmentations and Multicuts
4 Outer Relaxation of the Cycle Polytope
5 Lagrangian Decomposition
5.1 Constrained Reparameterization
6 Bound Maximization Along Subgradients
7 Rounding Heuristic and Interpretation
7.1 Decoding Heuristic: Iterative Construction
8 Experiments
8.1 Berkeley Segmentation Data Set
8.2 Correlation Clustering in Non-planar Graphs
9 Discussion
References
Learning from Imbalanced Data Using Ensemble Methods and Cluster-Based Undersampling
1 Introduction
2 Related Work
3 Proposed Algorithms
3.1 Undersampling Based on Clustering and K-Nearest Neighbour
3.2 Undersampling Based on Clustering and Ensemble Learning
4 Experiments and Results
4.1 Evaluation Criteria
4.2 Datasets and Experimental Settings
4.3 Results and Analyses
5 Conclusion and Future Work
References
Data Streams and Sequences
Prequential AUC for Classifier Evaluation and Drift Detection in Evolving Data Streams
1 Introduction
2 Evaluating Data Stream Classifiers
3 Prequential AUC
4 Drift Detection Using AUC
5 Experiments
5.1 Datasets
5.2 Results
6 Conclusions
References
Mining Positional Data Streams
1 Introduction
2 Related Work
3 Efficiently Finding Similar Movements
3.1 Representation
3.2 Approximate Dynamic Time Warping
3.3 An N-Best Algorithm
3.4 Distance-Based Hashing
4 Frequent Episode Mining for Positional Data
4.1 Counting Phase
4.2 Generation Phase
5 Empirical Evaluation
5.1 Positional Data
5.2 Near Neighbour Search
5.3 Episode Discovery
6 Conclusion
References
Visualization for Streaming Telecommunications Networks
1 Motivation
2 Related Work
2.1 Visualization
2.2 top-k Itemsets
3 Streaming Simulation System
3.1 Components
3.2 Landmark Windows
3.3 Sliding Windows
3.4 top-k Networks
4 Case Study
4.1 Data Description
4.2 Sliding Windows Visualization
4.3 top-k Landmark Window
5 Conclusions
References
Temporal Dependency Detection Between Interval-Based Event Sequences
1 Introduction
2 Temporal Dependencies
2.1 Temporal Dependency Assessment
2.2 Significant Temporal Dependencies Selection
3 Discovery of Temporal Dependencies
4 Experimental Study
4.1 Quantitative Experiments
4.2 Case Study
5 Related Work
6 Conclusion
References
Applications
Discovering Behavioural Patterns in Knowledge-Intensive Collaborative Processes
1 Introduction
1.1 Motivation
2 Related Work
3 Behavioural Pattern
4 Methodology
4.1 Case Study: Collaborative Research Activity
4.2 Log Building
4.3 Hierarchical Clustering
5 Experiments
5.1 Discussion
6 Conclusions and Future Work
References
Learning Complex Activity Preconditions in Process Mining
1 Introduction
2 Representation
3 Learning
4 Computational Complexity Issues
5 Exploitation
6 Evaluation
7 Conclusions
References
Location Prediction of Mobile Phone Users Using Apriori-Based Sequence Mining with Multiple Support Thresholds
1 Introduction
2 Previous Work
3 Proposed Technique
3.1 Preliminaries
3.2 Apriori-Based Sequence Mining Algorithm with Multiple Support Thresholds (ASMAMS)
4 Evaluation
5 Experimental Results
6 Conclusion
References
Pitch-Related Identification of Instruments in Classical Music Recordings
1 Introduction
2 Audio Data
2.1 Parametrization
3 Classification with Random Forests
3.1 Instrument and Pitch Identification
3.2 Cleaning
3.3 Training of Random Forests
4 Results
5 Summary and Conclusions
References
Author Index

Content (PDF)

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

New Frontiers in Mining Complex Patterns

Description

More details

Other editions

Additional editions

Persons

Content

System requirements