| Summary: | 碩士 === 國立臺中教育大學 === 教育資訊與測驗統計研究所 === 106 === In order to connect with the internationalized society, multilingual people are getting more and more popular. Learning English has been an indispensable option for people who want to learn the second language. There are more and more methods of learning English. From the traditional teaching methods in the past to the current combination of technology and algorithm, they improve the effectiveness of learning.
The purpose of this study is to design a video recommendation system based on student’s responses and to figure out if students can improve their learning effectiveness after using this system. At the same time, the research question is whether the use of the system is more effective than traditional teaching to improve students’ learning performance.
The basis of this recommendation system is using the concept of Latent Semantic Analysis (LSA). The most relevant videos recommended to a student is after analyzing the questions of student’s wrong responses. Using HTML, PHP, and MySQL as tools to design this recommendation system website and database.
In the experiment, we separate students into two groups, experimental group and control group. In the experimental group, students learnt according to recommend videos. On the other hand, students in the control group taught by the teacher in a traditional grouping class. The results showed that experimental group’s average score of the post-test is statistically significantly higher than its pre-test. In addition, experimental group’s learning effective is also statistically significantly higher than the control group. Therefore, the video recommendation system can help students to learn English effectively.
|
|---|
