26th INTERNATIONAL CONFERENCE ON COMPUTERS IN EDUCATION, Metro Manila, Philippines, 26 - 30 November 2018, pp.669-674
This paper aims to develop an image recommendation system to support informal vocabulary learning for a context-aware learning environment called SCROLL (System for Capturing and Reminding of Learning Log). The idea is based on ubiquitous learning logs analysis, image (contextual and non-contextual) feature analysis, and information visualization techniques. We designed a model based on which the proposed Feature-based Context-specific Appropriate Image (FCAI) recommendation system will be developed. The purpose of recommending FCAI images is to assist language learners in informal learning of foreign vocabulary. Those images are recommended by analyzing 25350 ubiquitous learning logs (i. e. a learner's ethnographic information, learning location, learning time, learning context, image information etc.) at the right time, right place and in the right learning context-basis. This paper also articulates the approaches for ubiquitous learning logs analysis, word-bank formation, image analysis methods, and algorithm development.