Integration of corpus linguistics and deep learning techniques for enhanced semantic-driven emotion detection on textual data.

Corpus linguistics provides an organized framework for examining a massive set of texts (corpora) to reveal structures, patterns, and interpretations in the language. Under the field of text-based emotion recognition (TER), corpus linguistics enables the extraction of linguistic features that inherently denote emotional expressions. With the application of annotated corpora encompassing emotionally labelled text, several studies train machine learning (ML) models for the identification and classification of emotions such as happiness, sad, anger, etc. The integration of linguistic corpus-based and data-driven approaches improves the accuracy and depth of emotion recognition, posing valuable understandings for applications in sentiment analysis (SA), social media monitoring, mental health assessment, and human-computer interaction (HCI). Therefore, this paper proposes an Emotion Detection in Text via Integrated Word Vector Representation and Multi-Model Deep Neural Networks (EDTIWVR-MDNN) approach. The EDTIWVR-MDNN approach intends to develop an effective textual emotion recognition and analysis to enhance sentiment understanding in natural language. Initially, the text pre-processing stage involves various levels for extracting the relevant data from the text. Furthermore, the word embedding process is performed by the fusion of the term frequency-inverse document frequency (TF-IDF), bi-directional encoder representations from transformers (BERT), and the GloVe technique to transform textual data into numerical vectors. For the classification process, the EDTIWVR-MDNN technique employs a hybrid of an attention mechanism-based temporal convolutional network and a bi-directional gated recurrent unit (AM-T-BiG) model. The experimental evaluation of the EDTIWVR-MDNN methodology is examined under the Emotion Detection from Text dataset. The comparison analysis of the EDTIWVR-MDNN methodology portrayed a superior accuracy value of 99.26% over existing approaches.