【摘要】：為了更好的保障網(wǎng)絡(luò)信息的安全穩(wěn)定性,需要進(jìn)行海量網(wǎng)頁掛碼信息自動采集方法研究,但是采用當(dāng)前方法進(jìn)行網(wǎng)頁掛碼信息自動采集時,無法構(gòu)造網(wǎng)頁掛碼信息的高維特征空間,存在網(wǎng)頁掛碼信息自動采集精度低的問題。為此,提出一種基于改進(jìn)神經(jīng)網(wǎng)絡(luò)的海量網(wǎng)頁掛碼信息自動采集方法。上述方法先采用神經(jīng)網(wǎng)絡(luò)對海量網(wǎng)頁掛碼信息樣本進(jìn)行標(biāo)準(zhǔn)化,得到信息特征的模糊隸屬度函數(shù),利用梯度優(yōu)化法進(jìn)行網(wǎng)絡(luò)訓(xùn)練,將最小二乘支持向量機(jī)參數(shù)編碼定義為蝙蝠個體,并以海量網(wǎng)頁掛碼信息自動采集有效性作為參數(shù)目標(biāo)優(yōu)化函數(shù),通過模擬蝙蝠飛行過程搜索到最小二乘支持向量機(jī)最優(yōu)參數(shù),以此為依據(jù)完成對海量網(wǎng)頁掛碼信息自動采集。仿真證明,所提方法信息采集精度較高,可以為保障網(wǎng)絡(luò)信息的安全穩(wěn)定性提供可行的依據(jù)。
[Abstract]:In order to better guarantee the security and stability of network information, it is necessary to study the automatic collection method of massive web page hanging code information, but when the current method is used to automatically collect the web page code information, It is impossible to construct the high dimensional feature space of web page code information, which has the problem of low precision of automatic collection of web page code information. In this paper, an improved neural network based method for automatic collection of massive web page coding information is proposed. Firstly, the neural network is used to standardize the information samples of massive web pages, and the fuzzy membership function of the information features is obtained, and the gradient optimization method is used to train the network. The least-squares support vector machine (LS-SVM) parameter coding is defined as bat individual, and the validity of massive webpage information collection is taken as the optimization function of parameter target, and the optimal parameters of LS-SVM are obtained by simulating bat flight process. Based on this, the automatic collection of massive web page code information is completed. Simulation results show that the proposed method has a high accuracy and can provide a feasible basis for ensuring the security and stability of network information.
【作者單位】： 西北師范大學(xué)計算機(jī)科學(xué)與工程學(xué)院;
【分類號】：TP393.08
，

本文編號：2299429