A Study on the macro-structure of Japanese text and constructing a database for teaching Japanese language

点击翻译按钮获取中文摘要

Recently, the need for teaching Japanese as a foreign language is increasing and the tendency to use Japanese to communicate through the internet and among the international community is also in great demand. But the teaching materials in reading is still in the stage of developing and it takes quite a time for us to be able to produce a good material or Japanese text books.The process of reading and how to understand the words, phrases and sentences which appear in the text book is not an easy task. Therefore, learners of Japanese have great difficulties in reading Japanese materials. We find that it is easier for learners of Japanese to understand Japanese vocabularies and grammar relatively. What we think is the most difficult thing in Japanese language is to learn how to read from the context, and to find out the macro-structure of the text.This research investigated the process of reading comprehension by using the methodology of Cognitive Science. We used a computer to enable us … More to know how the reading comprehension system has been constructed. The hardware used in this research is Macintosh and the software is HyperCard. We used " Shukyoteki-Kodo to Shukyo-ishki (Religious Behavior and Religious Consciousness)" as our reading article.The purpose of this system was to make learners understand the macro-structure of Japanese text and to investigate the process of reading. We provided 13 cards of HyperCard stacks which contain the text of "Shukyoteki-Kodo to Shukyo-ishki (Religious Behavior and Religious Consciousness)."The stacks consist of the Yomi 1 (Reading 1) and the Yomi 2 (Reading 2). For our data, we had more than 200 non-native Japanese students in advance level and Japanese students as our examinees.Next, we had the examinees read 13 cards and choose 13 key sentences from Yomi 1 and Yomi 2, then, we had them make a summary of the text from the 13 key sentences given. Then, we automatically in-put all the key sentences chosen by the examinees in the computer. Moreover, we also had them predict what would the content be in the next card.For this research, we found that non-native examinees tend to pay too much attention to words, phrases and/or micro-structure of the sentences. Consequently, they were misled from the key sentences and misunderstand the hierarchical structure of the text. Less