Multiple-Emotion Recognition in Gujarati Language Using Natural Language Processing

Understanding emotions when we talk to computers is essential and is an important aspect of human–computer interaction, and it has been widely researched for the English language. However, the problem of emotion recognition in native languages other than English has received less attention. This paper aims to develop and test a robust multiple-emotion (‘joy’, ‘surprise’, ‘love’, ‘sadness’, ‘fear’, and ‘anger’) detection model for the Gujarati language using natural language processing techniques. A dataset of labelled English tweets with emotions was translated into Gujarati using machine translation, and then classification algorithms were applied. The results of this study suggest that it is possible to build a robust emotion recognition model for Gujarati language and provide a foundation for future research in this area.