Description |
1 online resource (310 p.) |
Series |
Communications in Computer and Information Science ; 1943 |
|
Communications in computer and information science ; 1943.
|
Contents |
Intro -- Preface -- Organization -- Contents -- Research Track -- Random Padding Data Augmentation -- 1 Introduction -- 2 Related Work -- 2.1 Approaches to Improve Accuracy of CNNs -- 2.2 Padding in CNN -- 3 Random Padding for CNN -- 3.1 Random Padding Operation -- 3.2 Validation Method for Position Information Reduction in CNNs -- 3.3 Construct CNN with Random Padding -- 4 Evaluation of Position Information in CNNs -- 4.1 Dataset and Evaluation Metrics -- 4.2 Architectures and Settings -- 4.3 Comparison and Evaluation -- 5 Evaluation of Random Padding -- 5.1 Dataset and Evaluation Metrics |
|
5.2 Experiment Setting -- 5.3 Classification Accuracy on Different CNNs -- 5.4 Classification Accuracy on Different CNNs -- 6 Conclusions -- References -- Unsupervised Fraud Detection on Sparse Rating Networks -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 3.1 Problem Definition -- 3.2 Metrics -- 4 Methodology -- 4.1 Unsupervised Learning -- 4.2 Alleviating Network Sparsity -- 4.3 Avoiding Overfitting -- 4.4 Proposed Algorithm: FD-SpaN -- 4.5 Time Complexity Analysis -- 5 Experiments -- 5.1 Experiment Setup -- 5.2 Effectiveness (RQ1) -- 5.3 Training Percentage (RQ2) |
|
5.4 Linear Scalability (RQ3) -- 5.5 Ablation Study (RQ4) -- 6 Conclusion -- References -- Semi-supervised Model-Based Clustering for Ordinal Data -- 1 Introduction -- 2 Model Formulation -- 2.1 Proportional Odds Model -- 2.2 Proportional Odds Model with Clustering -- 2.3 Likelihood -- 3 Model Fitting -- 3.1 The Expectation Step (E-Step) -- 3.2 The Maximization Step (M-Step) -- 4 Simulation Study -- 5 Conclusions -- References -- Damage GAN: A Generative Model for Imbalanced Data -- 1 Introduction -- 2 Related Work -- 2.1 GAN -- 2.2 CNN -- 2.3 DCGAN -- 2.4 Contrastive Learning -- 2.5 ContraD GAN |
|
3 Methodology -- Damage GAN -- 4 Experimental Results -- 4.1 Datasets -- 4.2 Evaluation Metrics -- 4.3 Results -- 5 Conclusion -- References -- Text-Conditioned Graph Generation Using Discrete Graph Variational Autoencoders -- 1 Introduction -- 2 Related Work -- 2.1 Deep Learning on Graphs -- 2.2 Unconditioned Graph Generation -- 2.3 Conditioned Graph Generation -- 2.4 Text-to-Graph Problems -- 3 Model -- 3.1 Graph Vector-Quantized Variational Autoencoder -- 3.2 Transformer Decoder -- 4 Datasets -- 4.1 Graph-Text Paired Datasets -- 4.2 Graph Datasets -- 5 Experiments -- 5.1 Evaluation Metrics |
|
5.2 Results -- 6 Conclusion -- References -- Boosting QA Performance Through SA-Net and AA-Net with the Read+Verify Framework -- 1 Introduction -- 2 Related Work -- 3 Problem Definition -- 3.1 QA Task -- 3.2 Identifying the Presence or Absence of Answers -- 4 Method -- 4.1 Model Architecture -- 4.2 SA-Net and AA-Net -- 4.3 Ensemble Model Inference -- 4.4 Data Augmentation -- 5 Experiments -- 5.1 Evaluation -- 5.2 Experiment Results -- 6 Conclusion -- References -- Anomaly Detection Algorithms: Comparative Analysis and Explainability Perspectives -- 1 Introduction -- 2 Background |
Summary |
This book constitutes the proceedings of the 21st Australasian Conference on Data Science and Machine Learning, AusDM 2023, held in Auckland, New Zealand, during December 1113, 2023. The 20 full papers presented in this book were carefully reviewed and selected from 50 submissions. The papers are organized in the following topical sections: research track and application track. They deal with topics around data science and machine learning in everyday life. |
Notes |
2.1 Isolation Forest |
|
Includes author idnex |
|
Online resource; title from PDF title page (SpringerLink, viewed December 18, 2023) |
Subject |
Data mining -- Congresses
|
|
Data mining
|
Genre/Form |
Conference papers and proceedings
|
Form |
Electronic book
|
Author |
Benavides-Prado, Diana
|
|
Erfani, Sarah
|
|
Fournier-Viger, Philippe.
|
|
Boo, Yee Ling
|
|
Koh, Yun Sing, 1978-
|
ISBN |
9789819986965 |
|
9819986966 |
|