ABSTRACT. Cancer subtype recognition and feature selection are important problems in the diagnosis and treatment of tumors. Here, we propose a novel gene selection approach applied to gene expression data classification. First, two classical feature reduction methods including locally linear embedding (LLE) and rough set (RS) are summarized. The advantages and disadvantages of these algorithms were analyzed and an optimized model for tumor gene selection was developed based on LLE and neighborhood RS (NRS). Bhattacharyya distance was introduced to delete irrelevant genes, pair-wise redundant analysis was performed to remove strongly correlated genes, and the wavelet soft threshold was determined to eliminate noise in the gene datasets. Next, prior optimized search processing was carried out. A new approach combining dimension reduction of LLE and feature reduction of NRS (LLE-NRS) was developed for selecting gene subsets, and then an open source software Weka was applied to distinguish different tumor types and verify the cross-validation classification accuracy of our proposed method. The experimental results demonstrated that the classification performance of the proposed LLE-NRS for selecting gene subset outperforms those of other related models in terms of accuracy, and our proposed approach is feasible and effective in the field of highdimensional tumor classification.

Distributed denial of service (DDoS) attack is one of the most serious threats to the Internet The emergenceof distributed reflection denial of service (DRDoS) attacks has increased the harm of DDoSattacks. Aiming at the common DRDoS attacks such as Memcached, TFTP, NTP, SSDP, SNMPand Chargen in the network, a multi-class DRDoS attack detection method based on feature selectionis proposed. Through the analysis of the behavior and characteristics of attack, combined withprobability distribution of features and feature importance to obtain a feature subset of 24 features.When constructing XGBoost model, the input features are the feature subset obtained by the abovefeature selection, and the model outputs multi classification results. The selected features can betterreflect the characteristics of DRDoS attack and improve the detection performance of the model. Experimentalresults show that the feature subset obtained by this method has high precision in multiclassification against DRDoS attacks, and is better than the traditional methods such as support vectormachine and multi-layer perceptron. Feature selection not only reduces the processing time, butalso reduces the malicious traffic by 99.93%.