首页|Iqra University Researcher Discusses Research in Machine Learning (A Machine Lea rning-Based Framework with Enhanced Feature Selection and Resampling for Improve d Intrusion Detection)
Iqra University Researcher Discusses Research in Machine Learning (A Machine Lea rning-Based Framework with Enhanced Feature Selection and Resampling for Improve d Intrusion Detection)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Current study results on artificial in telligence have been published. According to news reporting out of Islamabad, Pa kistan, by NewsRx editors, research stated, "Intrusion Detection Systems (IDSs) play a crucial role in safeguarding network infrastructures from cyber threats a nd ensuring the integrity of highly sensitive data. Conventional IDS technologie s, although successful in achieving high levels of accuracy, frequently encounte r substantial model bias." Funders for this research include Princess Nourah Bint Abdulrahman University. The news journalists obtained a quote from the research from Iqra University: "T his bias is primarily caused by imbalances in the data and the lack of relevance of certain features. This study aims to tackle these challenges by proposing an advanced machine learning (ML) based IDS that minimizes misclassification error s and corrects model bias. As a result, the predictive accuracy and generalizabi lity of the IDS are significantly improved. The proposed system employs advanced feature selection techniques, such as Recursive Feature Elimination (RFE), sequ ential feature selection (SFS), and statistical feature selection, to refine the input feature set and minimize the impact of non-predictive attributes. In addi tion, this work incorporates data resampling methods such as Synthetic Minority Oversampling Technique and Edited Nearest Neighbor (SMOTE_ENN), Ada ptive Synthetic Sampling (ADASYN), and Synthetic Minority Oversampling Technique -Tomek Links (SMOTE_Tomek) to address class imbalance and improve t he accuracy of the model. The experimental results indicate that our proposed mo del, especially when utilizing the random forest (RF) algorithm, surpasses exist ing models regarding accuracy, precision, recall, and F Score across different d ata resampling methods."
NETL
NSTL
