Given the considerable importance of the English language as a common method of global communication, supporting and engaging learning environments for studying English as a second language are essential. In this study, we propose an interactive agent that considers differences in learners’ skill levels to help motivate learners and support independent study. The proposed agent is an AI chatbot that can vary the complexity of its natural language interactions as learners’ skills improve and engage in unstructured conversations with users. We first collected a dataset of English conversations among children and then acquired data on learners’ utterances of words and sentences using an interactive interface for skill evaluation. Based on these datasets, we generated a set of personalized conversational agents by using algorithms to evaluate speakers’ pronunciation and sentences. Finally, an expert in English language learning evaluated the interactive system to ensure that it could support learners with different levels of proficiency and that the application may be expected to help motivate students to learn. The results showed that the interactive teachable agent could help motivate students who may not otherwise be interested in learning by using a more personalized approach tailored to their skill level.
Instructors can now work with students to create various textbooks based on crowdsourcing. In particular, as feedback provided by students is essential for determining the quality and direction of classes, instructors should interact with students who are currently participating in classes by exchanging feedback. This paper proposes a block editing model that can reflect student feedback. The block editing model is an interactive e-textbook editing model that is dynamically updated based on the feedback provided by students in real time without modifying the structure of digital textbooks. In particular, in order for even non-developers who do not know web programming languages to be able to produce interactive digital textbooks easily, the authors developed an editor that could help implement them based on Blockly, a visual programming language. This paper enables instructors to improve the direction and quality of classes depending on the learning achievement of students and understanding based on feedback information provided by students and feedback analysis.