Team:
Sirbu Damian
Viola Daniele
Feature Reduction Method Comparison Towards Explainability and Efficiency in Cybersecurity Intrusion Detection Systems
Introduction
In the realm of cybersecurity, intrusion detection systems (IDS) detect and prevent attacks based on collected computer and network data. In recent research, IDS models have been constructed using machine learning (ML) and deep learning (DL) methods such as Random Forest (RF) and deep neural networks (DNN). Feature selection (FS) can be used to construct faster, more interpretable, and more accurate models. We look at three different FS techniques; RF information gain (RF-IG), correlation feature selection using the Bat Algorithm (CFS-BA), and CFS using the Aquila Optimizer (CFS-AO).
Models
There are 2 models that have to be considered: RF and DNN. Classification algorithms work by feeding a set of inputs through a model to output a final, categorical variable. The methods we will explore for this are RF and DNN. The basic building block for RF is the decision tree. Visualization is in fig. 1. The decision tree is a model that classifies data points by putting them through a series of binary decision boundaries and then assigning them to the same class as the majority of the training points within a final bucket.
The Bat algorithm is a population-based metaheuristics algorithm for solving continuous optimization problems. It’s been used to optimize solutions in cloud computing, feature selection, image processing, and control engineering problems.
Deep learning-based drug discovery for novel coronavirus 2019-nCoV/SARS-CoV-2
Introduction
The paper, published in the journal of Computational Biology and Chemistry in 2020, presents an intelligent system that can predict potential drug candidates for the treatment of COVID-19.
Solution
The researchers used deep learning models to analyze molecular structures of drugs and their interactions with the virus's proteins. They also used virtual screening techniques to identify compounds with the potential to bind to the virus's spike protein and block its entry into human cells.
he intelligent system identified several promising drug candidates, including Remdesivir, which has since been approved for the treatment of COVID-19 by the FDA. The study demonstrates the potential of intelligent systems in drug discovery and highlights the importance of collaboration between AI and medical researchers.
Conclusion
The authors suggest that the intelligent system could be used to screen existing drugs for repurposing, as well as to design new drugs with specific properties for COVID-19 treatment. This research provides a promising direction for future drug discovery efforts, which could ultimately lead to more effective treatments for COVID-19 and other diseases.
Sources:
https://arxiv.org/pdf/2303.12891.pdf
https://www.sciencedirect.com/science/article/pii/S1476927121001699
