Home // eKNOW 2015, The Seventh International Conference on Information, Process, and Knowledge Management // View article
Authors:
Toshiro Minami
Yoko Ohura
Keywords: Text Mining; Weight of Word for Grade Estimation; Text Analysis; Educational Data Mining; Lecture Data Analysis.
Abstract:
The eventual goal of our study is to extract useful knowledge which will help students with improving their learning performance. Towards this goal, we have studied the methods for extracting useful information about students' attitudes to the lectures they take from the lecture-related data. Such studies will contribute to capture the real status of university students, and will be able to make a very useful tool for student development in the future. In this paper, we challenge the problem of outcome/grade estimation from the text data that have been written by the students in a term-end questionnaire. First, we introduce a new concept for rating a keyword which will contribute to increasing of the grades of the students who use them. Then, we use the contribution rates and estimate the grades of students. We change the weights and compare the estimated grades with the original ones, so that we can find the optimal weights of the keywords in the proposed framework. Finally, by using the optimal rates of keywords, we compare the usage of keywords between high-graded and low-graded students.
Pages: 78 to 83
Copyright: Copyright (c) IARIA, 2015
Publication date: February 22, 2015
Published in: conference
ISSN: 2308-4375
ISBN: 978-1-61208-386-5
Location: Lisbon, Portugal
Dates: from February 22, 2015 to February 27, 2015