本文選題：擊鍵認(rèn)證 + 支持向量機(jī)�。� 參考：《西南財(cái)經(jīng)大學(xué)》2014年碩士論文

【摘要】：隨著社會(huì)的發(fā)展,科技進(jìn)步帶來的便利性越來越多體現(xiàn)在人們的生活中,例如電子商務(wù)將傳統(tǒng)商業(yè)理念和先進(jìn)的網(wǎng)絡(luò)技術(shù)的結(jié)合,完全顛覆了人們的生活方式,人們可以足不出戶完成交易。在進(jìn)行網(wǎng)上交易給我們帶來便利的同時(shí),僅僅憑靠賬戶和密碼完成的網(wǎng)上交易身份認(rèn)證顯然滿足不了現(xiàn)在的人們對(duì)網(wǎng)絡(luò)安全需求,因此身份識(shí)別技術(shù)正在受到越來越多的關(guān)注。傳統(tǒng)的身份識(shí)別技術(shù)安全性可靠程度較弱,于是可靠程度較高的生物特征識(shí)別技術(shù)漸漸進(jìn)入我們的視線。擊鍵認(rèn)證作為生物特征識(shí)別領(lǐng)域的重要一員,在近年來一直被廣泛研究。傳統(tǒng)的擊鍵認(rèn)證使用的都是擊鍵的持續(xù)時(shí)間以及按鍵偏移時(shí)間,然而在前人大量的實(shí)驗(yàn)結(jié)果看來,這樣簡(jiǎn)單的提取擊鍵生物特征是不足以描述用戶的擊鍵行為的。另外,用來驗(yàn)證實(shí)驗(yàn)?zāi)Ｐ驼鎸?shí)有效性的樣本數(shù)據(jù)是否能反應(yīng)用戶在真實(shí)場(chǎng)景下行為,還有待考究,這也就是在擊鍵認(rèn)證領(lǐng)域被研究了這么多后,依舊鮮有商業(yè)化產(chǎn)品問世的原因之一。 本文主要根據(jù)某公司的業(yè)務(wù)需求,通過對(duì)前人擊鍵認(rèn)證算法的改進(jìn),實(shí)現(xiàn)了完整的擊鍵輔助認(rèn)證系統(tǒng)。文章的主要內(nèi)容有： (1)根據(jù)軟件設(shè)計(jì)的規(guī)范,完成了擊鍵認(rèn)證系統(tǒng)架構(gòu)和接口設(shè)計(jì),另外,本系統(tǒng)一共實(shí)現(xiàn)了三種擊鍵識(shí)別的算法,其中主要是其中的SVM支持向量機(jī)算法與另外兩種算法做對(duì)比；系統(tǒng)流程設(shè)計(jì)與實(shí)現(xiàn),系統(tǒng)主要包括了數(shù)據(jù)采集,數(shù)據(jù)校驗(yàn)與預(yù)處理,提取特征,訓(xùn)練模型,模型預(yù)測(cè),模型存儲(chǔ)等功能模塊。數(shù)據(jù)上傳協(xié)議設(shè)計(jì),本文的數(shù)據(jù)采集量比較龐大,涉及面較廣,為了保護(hù)用戶自身隱私不受到侵犯,需要設(shè)計(jì)一種專用密碼傳輸協(xié)議；兩種基本擊鍵模型算法的實(shí)現(xiàn),為了驗(yàn)證本文提出的改進(jìn)算法,實(shí)現(xiàn)了貝葉斯分類以及Manhattan距離算法,用于檢驗(yàn)改進(jìn)算法模型的性能。 (2)重點(diǎn)論述了支持向量機(jī)算法的基本原理以及推導(dǎo)過程,算法中涉及到的基本概念以及算法運(yùn)行的完成流程,并對(duì)實(shí)際出現(xiàn)的工程問題進(jìn)行了規(guī)避和防控。其中通過實(shí)驗(yàn)淘汰了前人實(shí)驗(yàn)中認(rèn)為比較理想的按鍵特征,而根據(jù)具體實(shí)驗(yàn)結(jié)果選擇了差異更為顯著的時(shí)間特征。通過優(yōu)化支持向量機(jī)內(nèi)部的數(shù)據(jù)處理問題,提高了模型生成效率與模型預(yù)測(cè)效率。借助網(wǎng)格搜索的辦法,優(yōu)化模型參數(shù),達(dá)到了自動(dòng)優(yōu)化懲罰系數(shù)C和高斯核函數(shù)g的效果。通過實(shí)驗(yàn)結(jié)果,分析用戶樣本的特性,如密碼長(zhǎng)度,登陸頻率,登陸時(shí)間等與用戶賬戶風(fēng)險(xiǎn)之間關(guān)系,得出結(jié)果密碼長(zhǎng)度適中,能夠讓真實(shí)用戶不受影響,且能夠很好地防范入侵者的攻擊。 (3)比較改進(jìn)算法和另外兩種算法的優(yōu)劣和不足,并提出后續(xù)改進(jìn)思路。 在本文系統(tǒng)內(nèi)部測(cè)試階段,收集到了大量數(shù)據(jù),用戶完全在無(wú)感知的狀態(tài)下完成了擊鍵認(rèn)證實(shí)驗(yàn),能夠體現(xiàn)出用戶的擊鍵行為特征,實(shí)驗(yàn)數(shù)據(jù)真實(shí)有效。通過以這些真實(shí)數(shù)據(jù)為基礎(chǔ),將數(shù)據(jù)集分成訓(xùn)練集和測(cè)試集,用于構(gòu)建模型以及檢測(cè)模型算法的效果。在相同的訓(xùn)練集訓(xùn)練模型,測(cè)試集測(cè)試模型之后,Manhattan距離算法測(cè)試集正樣本的預(yù)測(cè)準(zhǔn)確率為82.87%,負(fù)樣本預(yù)測(cè)準(zhǔn)確率為81.67%；貝葉斯分類算法測(cè)試集正樣本的預(yù)測(cè)準(zhǔn)確率為87.78%,負(fù)樣本的預(yù)測(cè)準(zhǔn)確率為86.10%；支持向量機(jī)的測(cè)試集正樣本的預(yù)測(cè)準(zhǔn)確率為89.94%,負(fù)樣本的預(yù)測(cè)準(zhǔn)確率為95.80%。實(shí)驗(yàn)結(jié)果表明,本文的支持向量機(jī)算法,效果較為理想,達(dá)到了理想的識(shí)別精度。
[Abstract]:With the development of society , the convenience brought by scientific and technological progress is more and more reflected in people ' s life , such as the combination of traditional business concept and advanced network technology , which is not enough to describe the people ' s key behavior .

Based on the business needs of a company , this paper has realized the complete keystroke - assisted certification system by improving the key authentication algorithm of the former . The main contents of this paper are as follows :

( 1 ) According to the specification of software design , the structure and interface design of keystroke authentication system are completed . In addition , the system realizes three key recognition algorithms , in which SVM support vector machine algorithm is compared with the other two algorithms .
The system flow is designed and realized , the system mainly includes the function modules such as data collection , data checking and preprocessing , extracting features , training model , model prediction , model storage and so on . The data upload protocol is designed . The data acquisition in this paper is relatively large and the penetration is wide . In order to protect the privacy of the user , it is necessary to design a special cipher transmission protocol .
In order to validate the improved algorithm proposed in this paper , the Bayesian classification and the Manhattan distance algorithm are implemented to verify the performance of the improved algorithm model .

( 2 ) The basic principle of the support vector machine algorithm and the derivation process , the basic concept of the algorithm and the completion flow of the algorithm operation are discussed , and the time characteristic of the model generating efficiency and the model prediction efficiency is selected according to the experimental results .

( 3 ) Comparing the advantages and disadvantages of the improved algorithm and the other two algorithms , and putting forward the idea of improving the follow - up .

By taking these real data as the base , the data set is divided into training set and test set , which is used to construct the model and test model algorithm . After the same training set training model and the test set test model , the Manhattan distance algorithm test set positive sample is 82.87 % and the negative sample prediction accuracy is 81.67 % .
The prediction accuracy of positive samples was 87.78 % , and the accuracy of negative samples was 86.10 % .
The prediction accuracy rate of the test set positive samples of the support vector machine is 89.94 % , and the prediction accuracy of the negative samples is 95.80 % . The experimental results show that the support vector machine algorithm is ideal and the ideal recognition accuracy is achieved .
【學(xué)位授予單位】：西南財(cái)經(jīng)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TP393.08

【參考文獻(xiàn)】

相關(guān)期刊論文 前6條

1 李中獻(xiàn),詹榜華,楊義先;認(rèn)證理論與技術(shù)的發(fā)展[J];電子學(xué)報(bào);1999年01期

2 楊海燕;劉建成;道理;劉志華;;基于擊鍵動(dòng)力學(xué)和口令的雙因素用戶驗(yàn)證[J];信息技術(shù);2006年10期

3 朱建新,楊小虎,董金祥,曹哲新;生物認(rèn)證系統(tǒng)設(shè)計(jì)研究[J];計(jì)算機(jī)工程與應(yīng)用;2002年10期

4 鄭小霞;錢鋒;;高斯核支持向量機(jī)分類和模型參數(shù)選擇研究[J];計(jì)算機(jī)工程與應(yīng)用;2006年01期

5 高艷 ,管曉宏 ,孫國(guó)基 ,馮力;基于實(shí)時(shí)擊鍵序列的主機(jī)入侵檢測(cè)[J];計(jì)算機(jī)學(xué)報(bào);2004年03期

6 張浩;劉倩;;電子商務(wù)的信息安全問題探討與研究[J];商場(chǎng)現(xiàn)代化;2011年09期

，

本文編號(hào)：1966765