Hybrid Deep Learning Model for Emotion Recognition using Facial Expressions: Channelizing Employee Productivity

Abstract

Purpose – This paper aims to examine and predict various types of emotions in the humans using their facial expressions.

Design approach – The model developed in this work consists of two Convolutional Neural Networks (CNNs). The first CNN is used to analyse the primary emotion of the image as happy or sad. The second CNN is used to predict the secondary emotion of that image.

Findings – The overall results show that the proposed model is capable of predicting emotions using facial expressions better than the existing state of the art approaches.

Originality – The paper discusses a critical issue of predicting emotions as it can have potentially significant relevance in identifying an individual’s mental wellbeing in life in general and at workplace in particular. Organizations can also get benefitted as it can channelize employee productivity in right manner at work place.