Identity Verification Based on Keystroke Feature Using HMM
Ting Rui, Shen Chun-lin, Qi Tian, Jinlin Zhang · Jisuanji gongcheng · 2005
This paper proposes a new approach based on hidden Markov models (HMM). It analyzes keystroke feature extraction, observation encoding and HMM parameter training. To demonstrate the effectiveness of the proposed approach, it is compared against three best existing ones and reports promising results.