Incidental word learning tasks are widely adopted in pedagogical activities and self-paced learning processes, given their advantages of providing rich contexts and training on other language skills. While these tasks, diverse learners are often provided with the same contexts. For example, a cloze test with the same essay may be offered to all users in an e-learning system for learning target words. However, different learners may have varied expertise on and subjective preferences of many topics. Hence the provided unified learning context may be unfamiliar to some learners. The learning effectiveness is therefore likely to be negatively influenced. In response to a call to solve this problem, we propose in this paper a framework for word learning systems to automatically identify the context familiarity of individual learners based on their logs. A personalized approach to accurate recommendations of incidental word learning tasks is also devised according to the individual context familiarity. The results of our experimental studies on real participants show that the proposed framework and method promote significantly more effective word learning and increase the learning enjoyment greatly than conventional approaches with unified learning contexts.
|Publication status||Published - Nov 2015|
CitationXie, H., Zou, D., Wang, F. L., & Wong, T. L., & Wu, Q. (2015, November). Context-aware personalized courses search based on hybrid learner profile. Paper presented at the 23rd International Conference on Computers in Education (ICCE 2015): Transforming education in the big data era, The First World Hotel, Hangzhou, China.
- Context familiarity
- Incidental word learning
- Learner profile
- Task recommendation