Active learning for data streams: a survey

Machine Learning, Journal Year: 2023, Volume and Issue: 113(1), P. 185 - 239

Published: Nov. 20, 2023

Language: Английский

---

Empirical analysis of performance assessment for imbalanced classification

Machine Learning, Journal Year: 2024, Volume and Issue: 113(8), P. 5533 - 5575

Published: Jan. 23, 2024

Language: Английский

---

Evolving cybersecurity frontiers: A comprehensive survey on concept drift and feature dynamics aware machine and deep learning in intrusion detection systems

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 137, P. 109143 - 109143

Published: Aug. 22, 2024

Language: Английский

---

Improving GBDT performance on imbalanced datasets: An empirical study of class-balanced loss functions

Neurocomputing, Journal Year: 2025, Volume and Issue: unknown, P. 129896 - 129896

Published: March 1, 2025

Language: Английский

---

Class balancing diversity multimodal ensemble for Alzheimer’s disease diagnosis and early detection

Computerized Medical Imaging and Graphics, Journal Year: 2025, Volume and Issue: unknown, P. 102529 - 102529

Published: March 1, 2025

Language: Английский

---

Toward Privacy Preservation Using Clustering Based Anonymization: Recent Advances and Future Research Outlook

IEEE Access, Journal Year: 2022, Volume and Issue: 10, P. 53066 - 53097

Published: Jan. 1, 2022

With the continuous increase in avenues of personal data generation, privacy protection has become a hot research topic resulting various proposed mechanisms to address this social issue. The main technical solutions for guaranteeing user's are encryption, pseudonymization, anonymization, differential (DP), and obfuscation. Despite success other solutions, anonymization been widely used commercial settings preservation because its algorithmic simplicity low computing overhead. It facilitates unconstrained analysis published that DP latest techniques cannot offer, it is mainstream solution responsible science. In paper, we present comprehensive clustering-based (CAMs) have recently preserve both utility publishing. We systematically categorize existing CAMs based on heterogeneous types (tables, graphs, matrixes, etc.), an up-to-date, extensive review metrics their evaluation. discuss superiority effectiveness over traditional mechanisms. highlight significance different paradigms, such as networks, internet things, cloud computing, AI, location-based systems with regard preservation. Furthermore, representative compromise individual privacy, rather than safeguarding it. Finally, challenges applying CAMs, suggest promising opportunities future research. To best our knowledge, first work cover current involving paradigms.

Language: Английский

---

Nonstationary data stream classification with online active learning and siamese neural networks✩

Neurocomputing, Journal Year: 2022, Volume and Issue: 512, P. 235 - 252

Published: Sept. 12, 2022

Language: Английский

---

An adaptive imbalance modified online broad learning system-based fault diagnosis for imbalanced chemical process data stream

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 234, P. 121159 - 121159

Published: Aug. 14, 2023

Language: Английский

---

Resampling strategies for imbalanced regression: a survey and empirical analysis

Artificial Intelligence Review, Journal Year: 2024, Volume and Issue: 57(4)

Published: March 4, 2024

Abstract Imbalanced problems can arise in different real-world situations, and to address this, certain strategies the form of resampling or balancing algorithms are proposed. This issue has largely been studied context classification, yet, same problem features regression tasks, where target values continuous. work presents an extensive experimental study comprising various predictive models, wich uses metrics capture important elements for user evaluate model imbalanced data context. It also proposes a taxonomy approaches based on three crucial criteria: model, learning process, evaluation metrics. The offers new insights into use such strategies, highlighting advantages they bring each model’s indicating directions further studies. code, information related experiments performed herein be found GitHub: https://github.com/JusciAvelino/imbalancedRegression .

Language: Английский

---

Analyzing the Performance and Efficiency of Machine Learning Algorithms, such as Deep Learning, Decision Trees, or Support Vector Machines, on Various Datasets and Applications

Deleted Journal, Journal Year: 2024, Volume and Issue: 3(2), P. 126 - 136

Published: Jan. 15, 2024

This research endeavors to comprehensively evaluate and compare the performance of three prominent machine learning algorithms—Deep Learning (DL), Decision Trees (DT), Support Vector Machines (SVM)—across a spectrum diverse datasets applications. The study is driven by specific objectives, including quantitative analysis accuracy, precision, recall, F1 Score for each algorithm discern their nuanced strengths weaknesses in varied contexts. Additionally, aims investigate impact algorithmic factors, such as complexity interpretability, on these models. By exploring trade-offs associated with sophisticated models interpretable alternatives, contributes valuable insights selection criteria. Another crucial objective analyze effect dataset characteristics, size, complexity, class imbalance, behavior, offering into challenges posed different potential strategies addressing issues imbalances biases. Furthermore, seeks assess generalization capabilities algorithms across application domains, encompassing image classification, natural language processing, numerical prediction. Lastly, delves ethical considerations, specifically focusing bias assessment transparency measures decision-making. emphasizing responsible AI deployment, addresses biases ensures through availability code datasets. structured approach objectives provides clear roadmap an in-depth investigation performance, influential considerations deployment algorithms.

Language: Английский

---