Machine Learning Methodologies for Preventing Malware Obfuscation

Carletti, Vincenzo; Saggese, Alessia; Foggia, Pasquale; Greco, Antonio; Vento, Mario

doi:10.1007/978-3-031-21940-5_6

Cited by 2 publications

(2 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a parallel development, the SkipGNN model \cite{huang2020skipgnn} adopts a dual architecture of Graph Neural Networks (GNN) to aggregate information and make predictions regarding drug-drug interactions. Meanwhile, the TriVec model pioneers a novel approach to knowledge graph embedding, where embedded vectors are deployed to predict polypharmacy side effects by modeling the available data as a knowledge graph [34]. In a more recent innovation [35], the GFAN model introduces a graph feature attention network designed to generate interpretable predictions concerning polypharmacy side effects [36].…”

Section: Recent Developments In Predicting Polypharmacy Side Effectsmentioning

confidence: 99%

Predicting Big data Drug Interactions and associated side effects by Using Artificial Neural Networks (ANN) over Traditional Graph Convolutional Networks (GCNs)

IBRAHIM,

Mohammed,

DARA

2024

Preprint

View full text Add to dashboard Cite

Rapid advances in machine learning have enabled the prediction of complex drug-drug interactions (DDIs) and associated harmful effects. This study aims to develop a neural network model that can predict drug-drug interactions (DDIs) for various side effects. Our study intends to create a reliable and easy-to-understand tool that can transform pharmaceutical research and healthcare by lowering polypharmacy risks. Our method begins with the careful selection and compilation of large datasets on medicine combinations, side effects, drug-side effect connections, and drug-protein interactions. We use an adjacency matrix to establish a drug-protein network. Then, we use PCA to shrink the network. Using artificial neural networks, our neural network is designed for binary categorization. This model is rigorously trained, validated, and tested using performance metrics to ensure its strength and adaptability. Our model has remarkable accuracy, with AUC-ROC scores of 98.67% for certain interactions. Reading and handling structured input is a major advantage of Artificial Neural Networks (ANNs) over Graph Convolutional Networks (GCNs). The findings demonstrate our approach's versatility in pharmaceutical research and healthcare, including medication development and real-time clinical decision help. To conclude, this study advances DDI prediction and management. Resilience, comprehensibility, and accuracy make the model a flexible polypharmacy solution. By solving DDI prediction and side effect control, our strategy might improve pharmaceutical research, patient safety, and healthcare results. This study shows how advanced machine learning may be used in pharmaceuticals.

show abstract

Section: Recent Developments In Predicting Polypharmacy Side Effectsmentioning

confidence: 99%

Predicting Big data Drug Interactions and associated side effects by Using Artificial Neural Networks (ANN) over Traditional Graph Convolutional Networks (GCNs)

IBRAHIM,

Mohammed,

DARA

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…In [30], The authors addressed a consistency study against obfuscation performed on four CNNs, namely ResNet50, InceptionV3, VGG16, and MobileNet, which are frequently utilized for developing image-based malware classification systems. To that end, the authors proposed to retrain the CNN models using TL to classify malware from 9 distinct families using a well-known dataset benchmark.…”

Section: Related Workmentioning

confidence: 99%

A Novel Detection and Multi-Classification Approach for IoT-Malware Using Random Forest Voting of Fine-Tuning Convolutional Neural Networks

Atitallah

Driss

Almomani

2022

Sensors

View full text Add to dashboard Cite

The Internet of Things (IoT) is prone to malware assaults due to its simple installation and autonomous operating qualities. IoT devices have become the most tempting targets of malware due to well-known vulnerabilities such as weak, guessable, or hard-coded passwords, a lack of secure update procedures, and unsecured network connections. Traditional static IoT malware detection and analysis methods have been shown to be unsatisfactory solutions to understanding IoT malware behavior for mitigation and prevention. Deep learning models have made huge strides in the realm of cybersecurity in recent years, thanks to their tremendous data mining, learning, and expression capabilities, thus easing the burden on malware analysts. In this context, a novel detection and multi-classification vision-based approach for IoT-malware is proposed. This approach makes use of the benefits of deep transfer learning methodology and incorporates the fine-tuning method and various ensembling strategies to increase detection and classification performance without having to develop the training models from scratch. It adopts the fusion of 3 CNNs, ResNet18, MobileNetV2, and DenseNet161, by using the random forest voting strategy. Experiments are carried out using a publicly available dataset, MaleVis, to assess and validate the suggested approach. MaleVis contains 14,226 RGB converted images representing 25 malware classes and one benign class. The obtained findings show that our suggested approach outperforms the existing state-of-the-art solutions in terms of detection and classification performance; it achieves a precision of 98.74%, recall of 98.67%, a specificity of 98.79%, F1-score of 98.70%, MCC of 98.65%, an accuracy of 98.68%, and an average processing time per malware classification of 672 ms.

show abstract

Machine Learning Methodologies for Preventing Malware Obfuscation

Cited by 2 publications

References 41 publications

Predicting Big data Drug Interactions and associated side effects by Using Artificial Neural Networks (ANN) over Traditional Graph Convolutional Networks (GCNs)

Predicting Big data Drug Interactions and associated side effects by Using Artificial Neural Networks (ANN) over Traditional Graph Convolutional Networks (GCNs)

A Novel Detection and Multi-Classification Approach for IoT-Malware Using Random Forest Voting of Fine-Tuning Convolutional Neural Networks

Contact Info

Product

Resources

About