Neural Network-Based Undersampling Techniques

Arefeen, Md Adnan; Nimi, Sumaiya Tabassum; Rahman, M. Sohel

doi:10.1109/tsmc.2020.3016283

Cited by 40 publications

(9 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most of machine learning algorithms perform poorly on imbalanced datasets, and model will be completely biased toward majority instances, and it will ignore minority ones. Different techniques have been utilized to balance the imbalanced dataset, such as random undersampling [24, 61, 62], cluster undersampling [63, 64], and SMOTE technique [25, 28]. In this study, we developed a new undersampling algorithm namely One-SVM-US using One-class Support Vector Machine (SVM) to deal with imbalanced data.…”

Section: Methodsmentioning

confidence: 99%

Improving prediction of drug-target interactions based on fusing multiple features with data balancing and feature selection techniques

Khojasteh

Pirgazi

2022

Preprint

View full text Add to dashboard Cite

Predicting drug-target interaction (DTI) is an important research area in the field of drug discovery. It means identifying the interaction between chemical compounds and protein targets. Wet lab experiments to explore these interactions are expensive as well as time-consuming. On the contrary, a dry lab environment focusing more on computational methods of interaction prediction can be helpful to limit the search space for these experiments and give clues before developing a new medicine. This paper proposes a novel drug-target interaction prediction method called SRX-DTI. First, we extract various descriptors from protein sequences, and the drug is encoded as an FP2 molecular fingerprint. For handling the class imbalance problem, we propose the One-SVM-US technique to deal with imbalanced data. We also developed the FFS-RF algorithm, a forward feature selection algorithm, and coupled it with the random forest (RF) classifier to maximize the predictive performance. The forward feature selection algorithm adds new features to a set of selected features as far as the predictive power improves. This feature selection algorithm removes the irrelevant features to obtain the best optimal features. Finally, the balanced dataset with optimal features is given to the XGBoost classifier to identify DTIs. The experimental results demonstrate that our proposed approach SRX-DTI achieves significantly higher performance than other existing methods in predicting DTIs. The datasets and source code are available at: https://github.com/Khojasteh-hb/SRX-DTI .

show abstract

Section: Methodsmentioning

confidence: 99%

Improving prediction of drug-target interactions based on fusing multiple features with data balancing and feature selection techniques

Khojasteh

Pirgazi

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The model combined the advantages of a single multilayer perceptron classifier and C4.5 decision tree. Arefeen, MA et al [ 12 ] proposed two novel algorithms that employ neural network-based approaches to remove majority samples that are found to reside in the vicinity of the minority samples, thereby undersampling the former to remove (or alleviate) the imbalance issue.…”

Section: Related Workmentioning

confidence: 99%

Dual generative adversarial networks based on regression and neighbor characteristics

Jia,

Lu,

Shen

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

Imbalanced data is a problem in that the number of samples in different categories or target value ranges varies greatly. Data imbalance imposes excellent challenges to machine learning and pattern recognition. The performance of machine learning models leans to be partially towards the majority of samples in the imbalanced dataset, which will further affect the effect of the model. The imbalanced data problem includes an imbalanced categorical problem and an imbalanced regression problem. Many studies have been developed to address the issue of imbalanced classification data. Nevertheless, the imbalanced regression problem has not been well-researched. In order to solve the problem of unbalanced regression data, we define an RNGRU model that can simultaneously learn the regression characteristics and neighbor characteristics of regression samples. To obtain the most comprehensive sample information of regression samples, the model uses the idea of confrontation to determine the proportion between the regression characteristics and neighbor characteristics of the original samples. According to the regression characteristics of the regression samples, an index ccr (correlation change rate) is proposed to evaluate the similarity between the generated samples and the original samples. And on this basis, an RNGAN model is proposed to reduce the similarity between the generated samples and the original samples by using the idea of confrontation.

show abstract

“…A healthy number of intrepid researchers have applied oversampling [9][10][11][12][13][14], undersampling [15][16][17][18][19], and hybrid [20][21][22][23] preprocessing methods to restore balance to their training datasets. These methods are combined with feature classification methods to maximize benefits.…”

Section: A Data-level Mitigation Effortsmentioning

confidence: 99%

Hybrid Intrusion Detection System Based on Data Resampling and Deep Learning

Chen,

You,

Shiue

2024

IJACSA

View full text Add to dashboard Cite

The growth of the internet has advanced information-sharing capabilities and vastly increased the importance of global network security. However, because new and inconspicuous abnormal behaviors are nearly impossible to detect in massive network access environments, modern intrusion detection systems have identified a high rate of false-positive (FP) and false-negative (FN) attacks. To overcome this, this paper proposes a hybrid deep learning model that significantly mitigates the disadvantages of consistently imbalanced sample attack data. First, it resolves imbalanced data using random undersampling and synthetic minority oversampling techniques. Then, convolutional neural networks (CNNs) extract local and spatial features, and a transformer encoder extracts global and temporal features. The novelty of this combination increases recognition accuracy at the algorithm level, which is crucial to reducing FPs and FNs. The model was subjected to multiclassification testing on the NSL-KDD and CICIDS2017 benchmark datasets, and the results show that our model has higher classification accuracy and lower FP rates than state-ofthe-art intrusion detection models. Moreover, it significantly improves the detection rate of low-frequency attacks.

show abstract

Neural Network-Based Undersampling Techniques

Cited by 40 publications

References 33 publications

Improving prediction of drug-target interactions based on fusing multiple features with data balancing and feature selection techniques

Improving prediction of drug-target interactions based on fusing multiple features with data balancing and feature selection techniques

Dual generative adversarial networks based on regression and neighbor characteristics

Hybrid Intrusion Detection System Based on Data Resampling and Deep Learning

Contact Info

Product

Resources

About