Non-Acted Text and Keystrokes Database and Learning Methods to Recognize Emotions

The modern computing applications are presently adapting to the convenient availability of huge and diverse data for making their pattern recognition methods smarter. Identification dominant emotion solely based on text generated by humans is essential human–computer interaction. This work presents a multimodal text-keystrokes dataset associated learning identification human emotions hidden in small text. For this, 69 participants collected multiple scenarios. Stimuli induced through videos controlled environment. After stimuli induction, write reviews about given scenario an unguided manner. Afterward, keystroke in-text features extracted from dataset. These used with assortment identify short An accuracy 86.95% achieved fusing features. Whereas, 100% obtained pleasure-displeasure classes using fusion keystroke/text features, tree-based feature selection method, support vector machine classifier. present also compared four state-of-the-art techniques same task, where results suggest that proposal performs better terms accuracy.