Investigating the role of chatbot-based language tutors utilizing deep learning to facilitate English language acquisition in mobile applications.

This article focuses on the development and evaluation of a chatbot-based language tutor employing deep learning (DL) techniques to facilitate English language acquisition in mobile applications. It introduces the Efficient Pigeon Inspired fused Bidirectional Long Short-Term Memory (EPI-BiLSTM) model, which enhances intent classification and grammatical error correction by combining bidirectional LSTM networks with a pigeon-inspired optimization algorithm. Using a publicly available chatbot-based English learning dataset, the study applies preprocessing methods, Term Frequency Inverse Document Frequency (TF-IDF) for feature extraction, and back translation for data augmentation to address data scarcity. Experimental results demonstrate that EPI-BiLSTM outperforms traditional models in domain classification (80.5%), intent recognition (90.3%), entity recognition (75.2%), and average accuracy (81.3%), indicating its potential to improve personalized, interactive language learning and proofreading in mobile environments.