基于键盘和鼠标击键行为的用户身份识别

User Identification Based on Keyboard and Mouse Keystroke Behavior

针对用户信息的安全性和解决单一击键行为特征识别度不高的问题,提出一种基于键盘和鼠标的击键特征识别方法。该方法通过用户聊天和点击聊天窗口等采集键盘和鼠标击键行为特征,将加权贝叶斯和欧式距离算法结合,从而实现用户身份识别。实验结果表明,该算法在32名用户真实击键记录组成的数据集上错误接受率FAR和错误拒绝率FRR分别为2.64%和3.38%,随着对外界环境因素的控制,该方法的准确率还可以进一步提升。

Aiming at the security of user information and solving the problem of low feature recognition of single keystrokes,a keying recognition method based on keyboard and mouse is proposed. This method combines the weighted Bayesian and Euclidean distance algorithm by collecting keyboard and mouse keystrokes through user chat and click chat window,so as to realize user identi-fication. The experimental results show that the false acceptance rate FAR and the false rejection rate FRR of the algorithm are 2.64% and 3.38% respectively on the data set of 32 users' real keystroke records. With the control of external environmental factors, the accuracy of the method can be further improved.