Meta-ADD: A meta-learning based pre-trained model for concept drift active detection

Yu, Hang; Zhang, Qingyong; Li, Tianyu; Lu, Jie; Wen, Yonggang; Zhang, Guangquan

doi:10.1016/j.ins.2022.07.022

Cited by 55 publications

(16 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The design and implementation of distribution-oriented drift detection techniques involve evaluating and distinguishing data distributions from previous and more recent data sets within specified time frames. Substantial shifts in data distribution frequently serve as indicators of concept drift, prompting the need for model adjustments [25]. Various techniques can be employed to quantify data distributions, encompassing approaches like calculating the mean, measuring divergence, assessing information entropy, and utilizing metrics such as the Kullback-Leibler (KL) divergence.…”

Section: Distribution-oriented Concept Drift Detectionmentioning

confidence: 99%

A Drift-Oriented Adaptive Framework for Concept Drift Detection in Large-Scale Internet-of-Medical-Things Data Streams

Maheshwari,

Bt. Osman,

Daud

et al. 2024

Preprint

View full text Add to dashboard Cite

BackgroundThe rise of the Internet-of-Medical-Things (IoMT) and smart devices has led to a substantial increase in extensive data streams in the healthcare domain. The interconnected nature of medical devices introduces dynamic and evolving data patterns. However, this dynamism poses a significant challenge known as Concept Drift, particularly crucial in the medical field. Concept Drift reflects the inherent instability in data patterns over time. In medical applications, this challenge intensifies as sensors must seamlessly transition from general healthcare monitoring to handling critical situations like emergency ICU operations. The complexity deepens due to imbalanced data distributions inherent in e-health scenarios.MethodsThe study introduces an Adaptive Ensemble Framework (AEF-CDA) designed to detect and adapt to concept drift in large-scale medical data streams from IoMT. The framework incorporates adaptive data preprocessing, a novel drift-centric adaptive feature selection approach, the learning of base models, and the selection of models adapted to concept drift. Additionally, an online ensemble model is integrated to enhance concept drift adaptation.ResultsThe proposed AEF-CDA framework is evaluated using three public IoMT and IoT datasets. The experimental results demonstrate its superiority over contemporary methods, achieving a remarkable accuracy of 99.64% with a precision of 99.39%. These metrics surpass the performance of other approaches in the simulation.ConclusionIn conclusion, the research presents a robust solution in the form of the adaptive ensemble framework (AEF-CDA) to effectively address the challenges posed by concept drift in IoMT data streams. The demonstrated high accuracy and precision underscore the framework’s efficacy, highlighting its potential significance in the dynamic landscape of medical data analysis.

show abstract

Section: Distribution-oriented Concept Drift Detectionmentioning

confidence: 99%

A Drift-Oriented Adaptive Framework for Concept Drift Detection in Large-Scale Internet-of-Medical-Things Data Streams

Maheshwari,

Bt. Osman,

Daud

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Another stream learning method is presented by ref. [29]. This method is developed to deal with concept recurrence with clustering.…”

Section: Stream Classificationmentioning

confidence: 99%

Fair-CMNB: Advancing Fairness-Aware Stream Learning with Naïve Bayes and Multi-Objective Optimization

Badar,

Fisichella

2024

BDCC

View full text Add to dashboard Cite

Fairness-aware mining of data streams is a challenging concern in the contemporary domain of machine learning. Many stream learning algorithms are used to replace humans in critical decision-making processes, e.g., hiring staff, assessing credit risk, etc. This calls for handling massive amounts of incoming information with minimal response delay while ensuring fair and high-quality decisions. Although deep learning has achieved success in various domains, its computational complexity may hinder real-time processing, making traditional algorithms more suitable. In this context, we propose a novel adaptation of Naïve Bayes to mitigate discrimination embedded in the streams while maintaining high predictive performance through multi-objective optimization (MOO). Class imbalance is an inherent problem in discrimination-aware learning paradigms. To deal with class imbalance, we propose a dynamic instance weighting module that gives more importance to new instances and less importance to obsolete instances based on their membership in a minority or majority class. We have conducted experiments on a range of streaming and static datasets and concluded that our proposed methodology outperforms existing state-of-the-art (SoTA) fairness-aware methods in terms of both discrimination score and balanced accuracy.

show abstract

“…These techniques include error rate-based drift detection [23], data distribution-based drift detection [24], and multiple hypothesis testing drift detection [25]. Furthermore, the identification of the type of concept drift as well as the occurrence of concept drift have been studied [26]. As fake news is susceptible to trends, it is very important that the decision model is tolerant to concept drift.…”

Section: Concept Driftmentioning

confidence: 99%

Evaluation of Accuracy Degradation Resulting from Concept Drift in a Fake News Detection System Using Emotional Expression

2023

View full text Add to dashboard Cite

Fake news on social media has become a social problem. Fake news refers to false information that is deliberately intended to deceive people. Several studies have been conducted on automatic detection systems that reduce the damage caused by fake news. However, most studies address the improvements made in detection accuracy, and real-world operations are rarely discussed. As the contents and expressions of fake news change over time, a model with a high detection accuracy loses accuracy after a few years. This phenomenon is called concept drift. As most conventional methods employ word representations, these methods exhibit accuracy degradation resulting from changes in word fads and usage. However, methods using the sentiment information of words can identify inflammatory sentences, which is a characteristic of fake news, and may suppress performance degradation caused by concept drift. In this study, a model using vector representations obtained from an emotion dictionary was compared with a model using conventional word embedding. Subsequently, we verified the resistance of the model to performance degradation. The results revealed the method using sentiment representation is less susceptible to concept drift. Models and learning methods that can achieve both detection accuracy and resistance to accuracy degradation can enable further development of fake news detection systems.

show abstract

Meta-ADD: A meta-learning based pre-trained model for concept drift active detection

Cited by 55 publications

References 21 publications

A Drift-Oriented Adaptive Framework for Concept Drift Detection in Large-Scale Internet-of-Medical-Things Data Streams

A Drift-Oriented Adaptive Framework for Concept Drift Detection in Large-Scale Internet-of-Medical-Things Data Streams

Fair-CMNB: Advancing Fairness-Aware Stream Learning with Naïve Bayes and Multi-Objective Optimization

Evaluation of Accuracy Degradation Resulting from Concept Drift in a Fake News Detection System Using Emotional Expression

Contact Info

Product

Resources

About