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 I
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 I

Ioanna Miliou Nico Piatkowski Panagiotis Papapetrou(Editor)

Springer (Publisher)

Published on 15. April 2024

XVI, 268 pages

E-Book

PDF with digital watermarking

System requirements

978-3-031-58547-0 (ISBN)

€106.99incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Preface
Organization
Contents - Part I
Contents - Part II
Foundations of AI and ML
Tackling the Abstraction and Reasoning Corpus (ARC) with Object-Centric Models and the MDL Principle
1 Introduction
2 Related Work
3 Object-Centric Models
3.1 Mixing Patterns and Functions
3.2 Parsing and Generating Grids with a Grid Model
3.3 Predict and Describe Grids with Task Models
4 MDL-Based Model Learning
4.1 Description Lengths
4.2 Search Space and Strategy
4.3 Pruning Phase
5 Evaluation
6 Conclusion
References
RMI-RRG: A Soft Protocol to Postulate Monotonicity Constraints for Tabular Datasets
1 Introduction
2 Related Work
3 Preliminaries and Notation
3.1 Rank Mutual Information
3.2 Relabeling
4 Main Direct Competitors
4.1 Subjective Approaches
4.2 Objective Approaches
5 RMI Tables and Required Relabelings Graphs
6 The RMI-RRG Protocol
7 Experimental Results
7.1 Breast Cancer
7.2 Car
7.3 CMC
7.4 Pasture
7.5 PIMA
7.6 Windsor
8 Conclusions
References
A Structural-Clustering Based Active Learning for Graph Neural Networks
1 Introduction
2 Related Work
3 Problem Formulation
3.1 Node Classification on Attributed Graphs
3.2 Graph Neural Networks (GNNs)
3.3 Active Learning Task for Graph Neural Networks
4 Proposed Method
4.1 Community Detection Using the SCAN Algorithm
4.2 Node Selection Based on PageRank
4.3 SPA Algorithm
5 Experiments
5.1 Experiment Settings
5.2 Dataset
5.3 Evaluation Metrics
5.4 Baselines Methods
6 Results
6.1 Experiment Results of SPA on GCN
6.2 Experiment Result of SPA on GraphSAGE
6.3 Complexity Analysis
7 Discussion and Conclusion
References
Multi-armed Bandits with Generalized Temporally-Partitioned Rewards
1 Introduction
2 Background and Related Work
3 Problem Formulation
4 Lower Bound on Regret
5 Proposed Algorithm and Regret Upper Bound
5.1 Proposed Algorithm: TP-UCB-FR-G
5.2 Regret Upper Bound of TP-UCB-FR-G
6 Experimental Results
6.1 Setting 1: Synthetic Environment
6.2 Setting 2: Spotify Playlists
7 Concluding Remarks and Future Work
References
GloNets: Globally Connected Neural Networks
1 Introduction
2 Notation and Model Definition
3 Related Work
4 Implementing GloNet
5 Experiments
6 Conclusions and Future Works
References
Mind the Data, Measuring the Performance Gap Between Tree Ensembles and Deep Learning on Tabular Data
1 Introduction
2 Preliminaries
2.1 Tabular Data
2.2 Tree Ensembles
2.3 Deep Learning
3 Related Work
4 Methodology and Design of Experiments
5 Results
5.1 Impact of Training Dataset Size
5.2 Feature Complexity
5.3 Explainability
6 Conclusions and Future Work
References
A Remark on Concept Drift for Dependent Data
1 Introduction
2 Problem Setup
2.1 A Probability Theoretical Framework for Concept Drift
2.2 Stochastic Processes
2.3 A Taxonomy of Change Detection in Data Streams
3 Consistency Property
3.1 Drift is not Non-Stationarity
3.2 Temporal Consistency
3.3 Measuring Consistency of a Noisy Stochastic Processes
4 Numerical Evaluation
4.1 Testing Stationarity
4.2 Evaluation of Method
5 Conclusion
References
Representation Learning
Variational Perspective on Fair Edge Prediction
1 Introduction
2 Related Work
3 Variational Fairness-Aware Node Embedding
3.1 Problem Set-Up
3.2 Definition of the Loss
3.3 Optimization of LEAVE
4 Experiments and Results
4.1 Edge Prediction Protocol
4.2 Evaluation Metrics
4.3 Baselines for Edge Prediction
4.4 Analysis of Results
5 Conclusion
References
Node Classification in Random Trees
1 Introduction
2 Related Work
2.1 Learning Probabilistic Graphical Models
2.2 Node Classification Using Graph Neural Networks
3 Method
3.1 Problem Formulation
3.2 Approach
3.3 GNN Design
3.4 Classifying Nodes
4 Evaluation
4.1 Dataset
4.2 Experiments
4.3 Results
5 Conclusion
References
Self-supervised Siamese Autoencoders
1 Introduction
2 Self-supervised Representation Learning
3 A Siamese Denoising Autoencoder
3.1 Motivation
3.2 Architecture
4 Experiments
4.1 Experimental Setup
4.2 Results
5 Related Work
6 Conclusion
References
Equivariant Parameter Sharing for Porous Crystalline Materials
1 Introduction
2 Related Work
3 Crystal Symmetries
4 Methods
5 Experiments
6 Discussion
References
Subgraph Mining for Graph Neural Networks
1 Introduction
2 Preliminaries
3 AutoGSN
3.1 Subgraph Mining
3.2 Selection
3.3 Counting
4 Experiments
5 Related Work
6 Conclusion
References
Applications
Super-Resolution Analysis for Landfill Waste Classification
1 Introduction
2 Related Work
2.1 Image Classification for Landfills Discovery
2.2 Image Quality Improvement
3 Methodology
3.1 Experimental Setup
3.2 Results
4 Conclusions
References
Predicting Performance Drift in AI Models of Healthcare Without Ground Truth Labels
1 Introduction
2 Related Work
3 Methods
3.1 Probabilistic Sources of Drift
3.2 Drift Detection Framework
4 Results
4.1 Simulated Data
4.2 UK Primary Care Covid-19 Data
5 Conclusions
References
An Interpretable Human-in-the-Loop Process to Improve Medical Image Classification
1 Introduction
2 Methodology
2.1 Dataset Description
2.2 Baseline Classifier Description
2.3 Concept-Based Interpretability
2.4 Concept Selection
2.5 Human-in-the-Loop Approach
3 Results and Discussion
3.1 Baseline Classifiers
3.2 Concept-Based Interpretability Results
3.3 Human-in-the-Loop Approach Results
4 Conclusion
References
Hybrid Ensemble-Based Travel Mode Prediction
1 Introduction
2 Related Works
3 Ensemble of Batch and Online Learners
3.1 Training of Online and Batch Learners with TMC Data Streams
3.2 Building an Ensemble of Batch and Online Learners
4 Results
4.1 Data Streams and Libraries
4.2 Experiments
4.3 Discussion
5 Conclusions
References
Natural Language Processing
Beyond Words: A Comparative Analysis of LLM Embeddings for Effective Clustering
1 Introduction
2 Related Work
3 Models and Algorithms
3.1 Clustering Algorithms
4 Numerical Experiments
4.1 Evaluation Metrics
4.2 Experimental Settings
4.3 Results and Discussion
5 Conclusion and Perspectives
References
Data Quality in NLP: Metrics and a Comprehensive Taxonomy
1 Introduction
1.1 Data Quality
2 Related Work
3 Taxonomy for Data Quality in NLP
3.1 Linguistic
3.2 Semantic
3.3 Anomaly
3.4 Classifier Performance
3.5 Diversity
4 Experimental Setup
5 Results and Discussion
6 Conclusion and Future Works
References
Building Brownian Bridges to Learn Dynamic Author Representations from Texts
1 Introduction
2 Related Works
3 BARL: Brownian Bridges for Author Representation Learning
3.1 Background
3.2 Using the Brownian Bridges
3.3 Variational Information Bottleneck
3.4 Learning Author Representations
3.5 Model Architecture of BARL
4 Experiments with BARL
4.1 Datasets
4.2 Parameter Settings and Competitors
4.3 Results in Authorship Attribution
4.4 Results in Document Dating
4.5 Results in Author Classification
4.6 Ablation Study
4.7 Qualitative Analysis
5 Conclusion
References
Automatically Detecting Political Viewpoints in Norwegian Text
1 Introduction
2 Related Work
2.1 Political Text Analysis
2.2 Domain- and Language-Specific LLMs
2.3 Masking Techniques
3 The nor-pvi Dataset
4 Encoder-Decoder Models
4.1 Training Datasets
4.2 Setup and Training
5 Experiments and Evaluations
6 Results and Discussions
7 Conclusion and Future Work
References
AHAM: Adapt, Help, Ask, Model Harvesting LLMs for Literature Mining
1 Introduction
2 Related Work
3 Experimental Data: Literature-Based Discovery Publications
4 Methodology
4.1 Domain-Adaptation via Sentence-Transformers and BERTopic
4.2 Prompt Engineering of LLMs to Design Topic Names
4.3 Assessing Adaptation Through Evaluation of Topic Naming
4.4 AHAM Heuristic
5 Quantitative Exploration of the AHAM Objective
6 Qualitative Evaluation
7 Conclusion and Further Work
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Advances in Intelligent Data Analysis XXII

Description

More details

Other editions

Additional editions

Content

System requirements