結(jié)合特征降維和NGO-CNN-BiLSTM的招考智慧平臺網(wǎng)絡(luò)異常流量檢測方法

打開文本圖片集

關(guān)鍵詞：招考智慧平臺；特征降維；卷積神經(jīng)網(wǎng)絡(luò)；雙向長短期記憶網(wǎng)絡(luò)；北方蒼鷹優(yōu)化算法DOI：10.15938/j. jhust. 2025.02.007中圖分類號：TP391 文獻標志碼：A 文章編號：1007-2683（2025）02-0064-09

Abstract：Inviewofthechalengesbroughtbycomputervirusesandnetwork maliciousattckstotheoperationandmaintenanceof therecruitmentintelligentplatfomsystem，thispaperproposesamethodtodetectabnoaltraffcofthercruitmentsmartplatfor basednfeaturedimensionalityreductioncombinedwithNorthernGoshawkOptimization（NGO）tooptimizeconvolutional neural network（CNN）andBi-directionalLong Short-Trm Memory（BiLSTM）.Byusing theKernelPrincipal ComponentAnalysis（KPCA） methodtoreducethedimensionalityofthenonlinearinformationcontaindinthedataset，therduceddataisusedasinputforthdep learningnetworkmodel.Then，theconvolutionkernelofCNNisoptimizedbyNGOtoobtaintheoptimalconvolutionkerel，andthe abnormatraffcwasdetectedbyBiLSTM.TheCIC-IDS-2017datasetisusedtoanalyzethetrainingandtestsamplesoftheitrusion detectionnetworkmodel，andtheacuracyandrainingtimeareimprovedcomparedwithother methods，whichconfirmsthefeasibility and effectiveness of this method.

Keywords：recruitmentintellgence platfom；featuredimensionalityeduction;ConvlutionalNeuralNetwork；Bi-directioalLong Short-Term Memory； Northern Goshawk Optimization

0 引言

招考智慧平臺具有更高的開放性，而開放性必然導(dǎo)致潛在的安全威脅，如何提升招考智慧平臺的入侵檢測系統(tǒng)的流量數(shù)據(jù)分析處理能力，并提高網(wǎng)絡(luò)攻擊檢測效率、異常數(shù)據(jù)分類精度和入侵檢測的自動化水平，是招考智慧平臺網(wǎng)絡(luò)安全面臨的核心和關(guān)鍵問題。（剩余13349字）