情緒對於敲擊鍵盤影響之研究

Abstract

情緒辨識科技在人機互動領域扮演著重要角色。近年來該領域正發展透過鍵盤敲擊辨識情緒的新技術。這項技術具備資料收集容易且在收集過程不易對終端使用者造成影響的優點。但由於目前只有較少的研究針對這項技術的原理及現象本身，本研究因此應運而生。本研究的研究目標為於實驗室執行受控實驗以驗證情緒對於鍵盤敲擊的影響。為達成研究目標，本研究包含兩個受控的人體試驗，來收集受測者在情緒被視覺刺激物及聽覺刺激物刺激下的鍵盤敲擊資料。資料分析的部分，透過雙因子情緒效價 (3) x 情緒喚起 (3）重複測量變異數分析來檢核實驗結果。本研究發現情緒對於鍵盤敲擊的效果相對於個體差異相當小，這項發現對於建構相關應用的研究者來說的意義是，若能在建立相關預測模型的過程中把個體差異納入自變項，將有效提升該模型的準確度。這項發現同時也解釋了為何過去在建構透過鍵盤辨識個體這類系統時，研究者鮮少提出情緒對此類系統辨識效能的影響。最後，因本研究中人體試驗所使用的情緒刺激物與問卷皆為當代學界通用標準且容易取得，這些實驗在往後的日子裡將能夠較容易的被重製。

Emotion recognition technology plays the essential role of enhancement in Human-Computer Interaction (HCI). In recent years, a novel approach for emotion recognition has been reported, which is by keystroke dynamics. The advantage of this approach is that the data used is rather unobtrusive and easy to obtain. There were only limited investigations about the phenomenon itself in previous studies. Hence, the goal of this dissertation is to examine the source of variance in keystroke typing patterns caused by emotions. To accomplish the goal, multiple human-subject experiments were conducted in laboratory. Two controlled experiments to collect subjects’ keystroke data in different emotional states induced by images and sounds were conducted. Two-way Valence (3) x Arousal (3) ANOVAs was used to examine the collected dataset. The results of the experiment are reported, and we found that the size of the emotional effect is small, compare to the individual variability. This finding suggests that the accuracy rate of emotion recognition technology could be further improved if personalized models are utilized. On the other hand, the finding also provides an explanation to that why real-world applications which authenticate the identity of users by monitoring keystrokes may not be interfered by the emotional states of users. Notably, the experiment was conducted using standard instruments and hence is expected to be reproducible.