.png){kind=icon}
返回
Keystroke dynamics identity verification - its problems and practical solutions
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cose.2004.02.004.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Password is the most widely used identity verification method in computer security domain. However, because of its simplicity, it is vulnerable to imposter attacks. Use of keystroke dynamics can result in a more secure verification system. Recently, Cho et at. (J Organ Comput Electron Commerce 10 (2000) 295) proposed autoassociative neural network approach, which used only the user's typing patterns, yet reporting a low error rate: 1.0% false rejection rate (FRR) and 0% false acceptance rate (FAR). However, the previous research had some limitations: (1) it took too long to train the model; (2) data were preprocessed subjectively by a human; and (3) a large data set was required. In this article, we propose the corresponding solutions for these limitations with an SVM novelty detector, GA-SVM wrapper feature subset selection, and an ensemble creation based on feature selection, respectively. Experimental results show that the proposed methods are promising, and that the keystroke dynamics is a viable and practical way to add more security to identity verification. (C) 2004 Elsevier Ltd. All rights reserved.
Keyword:
keystroke dynamics identify verification
novelty detection
autoassociative MLP
support vector machine
feature subset selection
ensemble based on feature selection
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
C
IF:
5.4
论文数:
4.6K
被引数:
1.4W
机构
暂无机构信息
引用论文
Adaptive Impedance Control to Enhance Human Skill on a Haptic Interface System自适应阻抗控制以增强触觉接口系统上的人类技能
OAAI
OAAI分享收藏Variation in aeolian environments recorded by the particle size distribution of lacustrine sediments in Ebinur Lake, northwest China.png)
SpringerPlusOAAI
SpringerPlusIF0
OAAI分享收藏分享收藏 没有更多内容
没有更多内容