Advances in Intelligent Data Analysis XXII

Name: Advances in Intelligent Data Analysis XXII | 22nd International Symposium on Intelligent Data Analysis, IDA 2024, Stockholm, Sweden, April 24-26, 2024, Proceedings, Part II
Brand: Springer
Price: 106.99 EUR
Availability: OnlineOnly

22nd International Symposium on Intelligent Data Analysis, IDA 2024, Stockholm, Sweden, April 24-26, 2024, Proceedings, Part II

Ioanna Miliou Nico Piatkowski Panagiotis Papapetrou(Herausgeber*in)

Springer (Verlag)

Erschienen am 15. April 2024

XVI, 278 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-031-58553-1 (ISBN)

106,99 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Organization
Contents - Part II
Contents - Part I
Temporal and Sequence Data
Kernel Corrector LSTM
1 Introduction
2 Related Work
2.1 LSTM
2.2 Data-Centric Time Series Forecasting
2.3 Corrector LSTM
3 Kernel Corrector LSTM
3.1 Training
3.2 Data Correction
4 Experimental Setup
4.1 Datasets
4.2 Evaluation Metrics
5 Results
5.1 Comparison with cLSTM
5.2 Comparison with LSTM
6 Conclusion
References
Unsupervised Representation Learning for Smart Transportation
1 Introduction
2 Related Work
3 Method
4 Experiments
4.1 Vehicle Acceleration Datasets
4.2 Encoder Details
4.3 Hyperparameter Tuning
4.4 Evaluation
5 Results and Discussion
6 Conclusion
References
T-DANTE: Detecting Group Behaviour in Spatio-Temporal Trajectories Using Context Information
1 Introduction
2 Background
3 Material and Methods
3.1 Problem Formulation
3.2 Affinity Learning Network
3.3 Graph Community Detection
4 Experiments
4.1 Datasets
4.2 Evaluation Metrics and Baselines
4.3 Results and Discussions
5 Conclusion
References
Statistical Learning
Backward Inference in Probabilistic Regressor Chains with Distributional Constraints
1 Introduction
2 Related Work
3 Our Method: Metropolis-Hasting Regressor Chains
4 Experiments
5 Results and Discussion
6 Conclusion
References
Empirical Comparison Between Cross-Validation and Mutation-Validation in Model Selection
1 Introduction and Related Work
2 Methods and Experimental Setup
3 Results
4 Discussion and Conclusion
References
Amplified Contribution Analysis for Federated Learning
1 Introduction
2 Prerequisites and Related Work
2.1 Federated Learning
2.2 Contribution Measures for Federated Learning
3 Amplified Deletion Analytics
3.1 Notation and Aggregation Masks
3.2 Leave-one-out and Alpha-Amplified Function
4 Experiments
4.1 Description
4.2 Results
4.3 Commentary
5 Conclusions
References
Data Mining
Monitoring Concept Drift in Continuous Federated Learning Platforms
1 Introduction
2 Related Work
2.1 Error-Based Drift Detection
2.2 Data-Based Drift Detection
3 Methodology
3.1 Concept Drift in CFL Platforms
3.2 Drift Detection
4 Preliminary Evaluation
4.1 Experimental Setup
4.2 CFL Performance Under Concept Drift
4.3 Detecting First Occurrence of Concept Drift
4.4 Detecting Best Performance Under Concept Drift
5 Conclusion, Limitation, and Future Work
References
S+t-SNE - Bringing Dimensionality Reduction to Data Streams
1 Introduction
2 Related Work
3 Streaming T-SNE (S+t-SNE)
3.1 Problem One - When to Start
3.2 Problem Two - Reduction of Space Fingerprint
3.3 Handling Drift
4 Experiments
4.1 Results
5 Conclusion
References
-DBSCAN: Augmenting DBSCAN with Prior Knowledge
1 Introduction
2 Related Work
3 Implementation
4 Evaluation
4.1 Synthetic Datasets
4.2 Real World Datasets
5 Conclusion and Future Work
References
Putting Sense into Incomplete Heterogeneous Data with Hypergraph Clustering Analysis
1 Introduction
2 Related Work
3 Hypergraph-Based Clustering Analysis Method
3.1 Step I: Hypergraph Construction
3.2 Step II: Transformation to Simple Graph
3.3 Step III: Cluster Integration and Analysis
3.4 Step IV: Deriving KPIs to Analyze Performance
4 Evaluation in Industrial Use-Case
4.1 Step I: Hypergraph Construction
4.2 Step II: Transformation to Simple Graph
4.3 Step III: Cluster Integration and Analysis
4.4 Step IV: Deriving KPIs to Analyze Performance
5 Conclusion
References
Optimization
Efficient Lookahead Decision Trees
1 Introduction
2 Related Work
3 Technical Background
3.1 Level-Two Specialized Algorithm
3.2 Level-Two Lookahead Information Gain
4 Less Greedy Decision Trees
5 Results
6 Conclusion
References
Learning Curve Extrapolation Methods Across Extrapolation Settings
1 Introduction
2 Related Work
3 Extrapolation Methods
4 Experimental Setup
5 Results and Discussion
6 Conclusion
References
Efficient NAS with FaDE on Hierarchical Spaces
1 Introduction
2 Fast DARTS Estimator
2.1 DARTS
2.2 Training a Chained Hierarchical Architecture Using DARTS
2.3 Deriving FaDE-Ranks on Hyper-Architecture
2.4 Joint Batch-Wise Pseudo Gradient Descent
3 Experiments
3.1 Validating FaDE-Ranks
3.2 NAS on Iterative FaDE-Ranks
4 Conclusion and Future Work
References
Investigating the Relation Between Problem Hardness and QUBO Properties
1 Introduction
2 Background
3 QUBO Formulations and Their Spectral Gaps
3.1 Kernel 2-Means Clustering
3.2 Simple Support Vector Machine Embedding
4 Experiments
4.1 Clustering
4.2 Support Vector Machine
5 Conclusion
References
XAI
Example-Based Explanations of Random Forest Predictions
1 Introduction
2 Modifying the Prediction Procedure of Random Forests
2.1 Random Forests
2.2 Modifying the Predictions
3 Empirical Investigation
3.1 Observing the Effective Number of Training Examples
3.2 Controlling the Number of Examples Used in the Predictions
4 Concluding Remarks
References
FLocalX - Local to Global Fuzzy Explanations for Black Box Classifiers
1 Introduction
2 Setting the Stage
3 Fuzzy Local to Global Explanation Framework
3.1 Local to Global Fuzzy Set Transformation
3.2 Global Fuzzy Set Theory Encoding
3.3 Global Explanation Theory Generation
4 Experiments
5 Conclusions and Future Work
References
Interpretable Quantile Regression by Optimal Decision Trees
1 Introduction
2 Related Work
3 Background
4 Quantile DL8.5
4.1 Simultaneous Tree Learning
4.2 Efficient Quantile Loss Computation
5 Experiments
5.1 Metrics
5.2 Results
6 Conclusion
References
SLIPMAP: Fast and Robust Manifold Visualisation for Explainable AI
1 Introduction
2 Problem Definition
2.1 SLIPMAP
2.2 Mapping from Covariates to the Target Variable
3 Algorithm
3.1 Computational Complexity
4 Datasets
5 Experiments
5.1 Predictions
5.2 Robustness
5.3 Local Explanation Comparison
5.4 Scaling
6 Conclusions and Future Work
References
A Frank System for Co-Evolutionary Hybrid Decision-Making
1 Introduction
2 Setting the Stage
3 A Frank System
4 Experiments
5 Conclusion
References
Industrial Challenge
Predicting the Failure of Component X in the Scania Dataset with Graph Neural Networks
1 Introduction
2 Methods
3 Exploratory Data Analysis
4 Data Preprocessing
5 Experiments
6 Conclusion
References
Towards Contextual, Cost-Efficient Predictive Maintenance in Heavy-Duty Trucks
1 Introduction
2 Problem Formulation
3 Methodology
3.1 Data Transformation
3.2 Predictive Model
4 Evaluation
5 Conclusion
References
Implementing Deep Learning Models for Imminent Component X Failures Prediction in Heavy-Duty Scania Trucks
1 Background
1.1 Predictive Maintenance
1.2 Contest Background Description
2 Analysis Method for Input Data Preparation
2.1 Exploring Raw Data
2.2 Preprocessing Raw Data
2.3 Preprocessing Train Labels
3 Modeling
3.1 Model Description
3.2 Summary of Model Performance
4 Reflection
5 Conclusion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken