Novel Text Emotion Classification Model Using Long Short-Term Memory Network

Abstract

Emotion has a great influence on human cognition, behavior, and communication. In last several years, there is a tremendous rise of users in social media platforms, due to which large amount of textual data got generated. Users continuously share their thoughts and sentiments, regardless of time or location. Furthermore, the growth of emotion detection mechanism for the social media platforms becomes truly remarkable. In this paper, one novel deep learning (DL)-based classification model is proposed which is underlined on the concepts of long short-term memory (LSTM) network for the classification of six different types of emotions such as sadness, anger, love, surprise, fear, and joy from English-based texts. The model is compared with nine distinctive state-of-the-art supervised learning models. Utilizing the 80%–20% train-test split, the proposed model has shown the higher classification accuracy of 90.28% in comparison to other compared models. Practically, this model can be used in chatbots, virtual assistants, and personalized applications during the interaction with users.

Received: 5 September 2024 | Revised: 20 February 2025 | Accepted: 4 September 2025

Conflicts of Interest

The author declares that he has no conflicts of interest to this work.

Data Availability Statement

The data that support the findings of this study are openly available in Kaggle at https://www.kaggle.com/datasets/praveengovi/emotions-dataset-for-nlp.

Author Contribution Statement

Rubul Kumar Bania: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing, Visualization.