• Turkan Mehraj Ismayilli

  Author

Knowledge of vocabulary plays an important role in building English language proficiency since it helps learners develop skills in terms of reading, writing, speaking, and listening. Yet, conventional vocabulary learning techniques usually offer fixed word lists and uniform learning material to everyone; hence, making them less personalized and failing to cope with the needs of individuals and fill their vocabulary gaps. In this regard, this research work suggests a novel vocabulary recommendation system based on reinforcement learning technology, which enables the delivery of personalized vocabulary learning support via recommendations that change depending on learners’ interactions and needs. The learner vocabulary profile is generated according to previous knowledge, difficult vocabulary items, retention skills, learning pace, and practice results of each learner. Further, on this basis, suitable vocabulary recommendations for learning and revision, along with relevant context examples and skill-based vocabulary practices, are offered to the learners. The efficacy of the system is estimated using an educational experiment lasting eight weeks in comparison to conventional vocabulary learning techniques. Results have shown that personalized vocabulary recommendations enhance the learning experience greatly. Personalized recommendations resulted in a 94.6% vocabulary improvement rate as opposed to an 82.4% vocabulary improvement rate of traditional learning. Personalized recommendations have been able to increase the vocabulary retention rate from 78.5% to 92.3% as well as the engagement level from 76.8% to 91.7%. It can be concluded from the above results that personalized vocabulary recommendations will enhance acquisition, retention, and application of vocabulary.

• PDF