Development and Evaluation of an Adaptive Mobile Learning System for Assisting English Reading and Listening

• Main Authors: Ching-kun Hsu, 許靜坤
• Other Authors: Chih-Kai Chang
• Format: Others
• Language: en_US
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/10140835239650643341

博士 === 國立臺南大學 === 數位學習科技學系博士班 === 100 === The dissertation developed and evaluated two mobile learning systems. One was used to support English reading. The other one was utilized to assist English listening. In terms of reading, the dissertation proposed a mobile learning system for adaptively assisting English reading. The system integrates the individual or collaborative reading annotation function with the reading material recommendation mechanism. Three experimental stages were conducted to evaluate the effects on individual reading and collaborative reading. Firstly, students read identical reading materials without reading recommendation mechanism in the experiment. They were assisted by the instant and shared reading annotation function. The experimental results indicated that those students who were heterogeneously grouped into twos, threes, and fours had significantly higher levels of reading comprehension than individual students, but those grouped into fives did not. Secondly, the dissertation evaluated the reading material recommendation function which was developed by an expert system method. In the experiment, the students read the material recommended by the system without using the reading annotation function. The accuracy of recommendation results is higher than ninety percent. The experimental results found the students had higher learning motivation when the students read adaptive reading materials. Thirdly, the dissertation assessed the integration of the mobile-assisted reading annotation function with the reading material recommendation mechanism. Only the students who had similar preferences and reading comprehension proficiency were recommended the same reading materials. The students should have the same reading materials when doing collaborative reading. Therefore, the collaborative group is homogeneous. The dissertation compared the reading comprehension of individual reading adaptively and non-adaptively with the reading comprehension of homogeneously collaborative reading. The experimental results showed that mobile-assisted individual or collaborative English reading adaptively had better reading comprehension and lower cognitive loads in comparison with the reading comprehension and cognitive loads of the non-adaptively individual reading. As for English listening, the dissertation confirmed the captions dependence of Taiwanese students by using eye-movement tracking machine to analyze the fixation targets of eyes when the students watched movies. The experimental results noted that more than seventy percent of the students are used to gaze and put their attention on captions. The dissertation processed the video captions by filtering and hiding the easier words but showing the harder words to carry out adaptive English listening learning on mobile devices. In addition, the experimental results found that the listening comprehension of the students using partial hidden captions was similar to the listening comprehension of the students using full captions. The hidden words avoid the students inputting information through reading text from eyes. The perception processing of ears becomes important so as to adapt the students to the features of reduced forms, assimilation, elision, and so on. On the contrary, unfamiliar vocabulary which was shown in the video captions so as to be read by the eyes of the students would assist students to recognizing the vocabulary they heard and prevent the students from confusing the vocabulary they heard with other words having similar pronunciation. Due to the limitation of human working memory, the dissertation calculated the percentage of the sentences which were filtered and had the left words fewer than or equal to 5, 7, 9 words. The test results of the 30 video samples showed that more than ninety-eight percent of sentences did not surpass the limitation of working memory after filtering the highest frequently used 2200 words from captions. Finally, the dissertation conducted an experiment to compare the effects of the identical caption-filtering and personalized caption-filtering on system usability, perceived satisfaction, enjoyment, and learning motivations. The experimental results indicated that the students had higher enjoyment and satisfaction with the personalized captions when watching videos to train English listening, resulting in higher intrinsic learning motivations. In conclusion, the dissertation proposed the integration of mobile-assisted reading annotation function with the reading material recommendation mechanism to adaptively assist students reading on mobile devices, and implemented a personalized mobile-assisted English listening learning system by an adaptive caption-filtering approach. The effects of the two mobile learning systems were also discussed in the dissertation.