Emotion Detection from Punjabi Text using Hybrid Support Vector Machine and Maximum Entropy Algorithm

Abstract: Emotion detection is the approach to detect the human emotion from different ways of images, text, videos and audios etc. We are focusing on the emotion detection from Punjabi text language. Vast amount of work has been done for the English language. In spite of lack of resources for Indian languages, work has been done for Telugu, Bengali and Hindi language. Punjab is one of Indian states and Punjabi is its official language. Punjabi is under resourced language. In this paper, we proposed a hybrid research approach for the emotion detection of the Punjabi text. The hybridization involves the concept of Support Vector machine and Maximum Entropy Algorithm. The dataset is considered from the Punjabi websites, news paper and various Punjabi blogs. The combined form of dataset is considered for the emotion detection. In this paper, we have considered six classes of emotions as Joy, Sadness, Fear, Surprise, Disgust and Anger. The goal of this research paper is to classify the considered dataset into the form of these six emotions.

Keywords: Emotion Detection, Punjabi dataset, Support Vector Machine, Maximum Entropy Algorithm, Emotions.