PiBAL-ML: A Machine Learning-Based Personalized Listening Platform for EFL Learners

Individuals exhibit variations in their needs, interests, and affective behaviours during the language learning process. Recognizing and accommodating these differences is crucial for optimizing language acquisition. Advancements in technology have provided numerous tools and opportunities to tailor language learning experiences to individual preferences, thereby facilitating a more effective learning journey. Regarding these rationales, this study aimed to propose a personalized language learning platform which was called Personal Interest Based Listening- Machine Learning (PiBAL-ML) and then determine its effect on English as Foreign Language learners’ listening skills. Thus, the study unfolded in two distinct phases: the design phase and the testing phase. The PiBAL-ML system, at its core, featured a diverse array of listening materials carefully curated to align with the learners' specific fields of interest. Furthermore, it included a variety of question types, and the number of questions was dynamically adjusted based on participants' performance and progress. The system was tested through explanatory sequential research design which included both quantitative and qualitative data gathered from university students. The quantitative results revealed that there was no significant difference in overall scores between the control group and the experimental group. However, experimental group outperformed control group in multiple-choice test questions. Lastly, the study highlighted that the learners were satisfied with the system as it presented different activities and texts related to their interests. Based on these findings, the study concluded that machine learning based personalization could enhance foreign language learning.