Name
Abstract | ||
|---|---|---|
This paper combines keystroke dynamics features with recently proposed keystroke acoustic features for user identification and authentication. Traditional keystroke dynamics uses temporal features. This paper explores the discriminative capability of the fusion of temporal features and acoustic features. This paper also explores the influence of filtering keystroke sound which turns out to be harmful for the performance of keystroke acoustic features. We collected a total of 824 samples from 7 subjects for experiments. 46 features including 38 acoustic features are extracted for the proposed user identification and authentication system. C-Support Vector Classification (C-SVC) and one-class Support Vector Machine (1-SVM) are applied for user identification and authentication respectively. Extensive experiments are carried out to verify the proposed system. Our work achieves 92.8% accuracy for user identification. The FRR (False Rejection Rate) and FAR (False Acceptance Rate) are only 12% and 11% for user authentication. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1109/MSN.2016.082 | 2016 12th International Conference on Mobile Ad-Hoc and Sensor Networks (MSN) |
Keywords | Field | DocType |
keystroke sound,keystroke dynamics,user identification,user authentication,biometrics | Mel-frequency cepstrum,Authentication,Computer science,Support vector machine,Keystroke logging,Filter (signal processing),Keystroke dynamics,Feature extraction,Speech recognition,Discriminative model | Conference |
ISBN | Citations | PageRank |
978-1-5090-5697-2 | 1 | 0.40 |
References | Authors | |
0 | 5 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Qianqian Zhou | 1 | 1 | 0.40 |
| Yanni Yang | 2 | 50 | 6.49 |
| Feng Hong | 3 | 214 | 16.08 |
| Yuan Feng | 4 | 47 | 3.52 |
| Zhongwen Guo | 5 | 116 | 13.32 |