Deep E-Cue:Systematic Deep Network for Multimodal Emotional Cues Recognition in Email Communication

Organizations require virtual communication because efficient communication is essential to successful company operations. Among the features of virtual communication typically email communication, which is textual and asynchronous, meaning that the sender and the recipient do not encounter it simultaneously. Email can be troublesome because it eliminates in-person interactions, which can distort our capacity to develop interpersonal cues that are essential for communicating, exchanging information, and establishing trust. The proposed research is aimed to investigate the adaptability of emotional cues in email communication. A deep learning model is built to extract the emotional cues in email considering multiple parameters like text, audio, video and context of communication. The proposed models achieve accuracy of 96.86%, precision of 97.44%, and sensitivity of 97.69<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> which is highest score when compared to other existing models.