Abstract
In the era of information explosion, individuals often need to read first-hand information online in English, which is a nonnative language of most of the people in the world. However, the lack of proficiency toward a non-native language makes it difficult for second-language readers to efficiently understand the contents on the English webpages, especially the long articles. In this work, we propose a novel reading interface to support English as second-language readers with adaptive display of machine translation (MT) using eye tracking, accompanied with sentence-level mapping using background colors. We conducted an experiment to investigate how different methods affect the second-language readers. We found that active translation could help second-language readers comprehend the English article without dazzling them. In addition, sentence-level mapping using background colors not only benefits the mapping between original sentences and their corresponding ones, but also alleviates the problems of line skipping and reading resumption.