英文摘要 |
The popularization of internet provides people with more opportunity for reading a great deal of digitized texts. From the perspective of Japanese pedagogy, the easy access of these authentic materials paves the way for bringing up Japanese learners' reading competence. However, it remains an unsettled question how to retrieve information that are useful for comprehension from considerable quantities of digitized texts and deliver it to learners effectively. Therefore, this study aims to propose a text mining based solution which assists learners of Japanese to quickly retrieve the gist of texts even without reading the whole context. It will analyze and extract the key words as well as their associated words from texts, then visualize them. In addition, Text-to-Speech technology will be employed to read the whole texts or designated words for the furtherance of reading comprehension.A built-in example sentence search module can present how a specified word is used in other contexts. Learners of Japanese take an advantage in learning kanji while this benefit brings a side effect at the same time. That is, many learners tend to mistake Chinese words for Japanese kanji. Therefore, it is necessary to clearly show the difference between Chinese words and Japanese kanji. A Sino-Japanese kanji comparison module is also developed to achieve this goal.It is hoped that the proposed reading assistance learning system will play a key role in information retrieval and improve learners' reading skills through versatile built-in modules. |