Mobile Computing for Language Testing in Blended Learning Environments: Combining Traditional and Digital Assessment Tools

This paper presents a novel approach to language testing by integrating mobile computing, gamification, and BERT (Bidirectional Encoder Representations from Transformers) into a unified platform for blended learning environments. The framework combines traditional assessments, such as essays and oral exams, with gamified digital tools like points, badges, levels, and leaderboards, creating an engaging, adaptive learning experience. BERT enhances task creation by generating contextually appropriate questions tailored to individual proficiency levels, ensuring that learners remain challenged without feeling overwhelmed. It evaluates structured and open-ended responses through advanced semantic and grammatical analysis, providing detailed feedback that highlights strengths and areas for improvement. Gamified elements, including storylines and challenges, immerse learners in real-world language scenarios, such as acting as a diplomat in cross-cultural communication tasks. The system’s adaptive features personalize task difficulty based on performance history, while rewards and rank progression sustain long-term engagement. By integrating multimodal inputs like text, speech, and gestures, the platform supports diverse learning styles and promotes accessibility. Performance metrics from this study demonstrate the platform’s superiority over traditional methods and alternative deep learning models, showcasing significant improvements in learner engagement, proficiency, and satisfaction. This innovative blend of gamification and advanced natural language processing holds transformative potential for language education, offering a dynamic, interactive, and student-centered approach that bridges classroom instruction and digital learning.