Text Emotion Recognition Based on Deep Learning and Attention with Whale Optimization Algorithm

Text Emotion Recognition (TER) represents one of the crucial tasks in Natural Language Processing (NLP), and it is considered to be highly essential for many applications, which include Social Network Analysis, Mental Health Assessment, and Public Opinion Monitoring. For improving the performance of TER, advanced deep learning (DL) model GRU-BiGRU-CNN-Att is presented in this research. It is a combination of Gated Recurrent Units (GRU) and Bi-directional Gated Recurrent Units (Bi-GRU) with a Convolutional Neural Network (CNN) as well as an attention mechanism (AM). Using a GRU, Bi-GRU, and CNN as part of deep feature extraction (FE) model, combined with an AM that gives important information different weights, the method that is proposed in the present work enhances the quality of the word vectors and leads to increasing sentiment analysis judgment accuracy. Whale Optimization Algorithm (WOA) that selects the most informative features for further enhancing the model, is used lastly for the optimization of feature selection process. After such optimization, those chosen features are trained on a Multi-Layer Perceptron (MLP), successfully combining machine learning (ML) and DL approaches for improving text emotion recognition. A varied corpus of tweets, sentences and dialogues had been used in the present work in order to assess the performance of the suggested model. With a 83.76% rate, the suggested approach had shown a notable accuracy increase