|
|
|
|
| LEADER |
00000cam a2200000 7i4500 |
| 001 |
0000100566 |
| 005 |
20220111093000.0 |
| 008 |
211207s2021 my eng |
| 040 |
|
|
|a UniSZA
|e rda
|
| 050 |
0 |
0 |
|a LB1044.87
|b .M64 2021
|
| 090 |
0 |
0 |
|a LB1044.87
|b .M64 2021
|
| 100 |
0 |
|
|a Mohd Fahmi bin Husin ,
|e author
|
| 245 |
0 |
0 |
|a Classification model of students' performance based on interaction on massive open online courses (MOOCs)
|c Mohd Fahmi bin Husin
|
| 264 |
|
0 |
|c 2021
|
| 300 |
|
|
|a xv, 108 leaves ;
|c 31 cm.
|
| 336 |
|
|
|a text
|2 rdacontent
|
| 337 |
|
|
|a unmediated
|2 rdamedia
|
| 338 |
|
|
|a volume
|2 rdacarrier
|
| 502 |
|
|
|a Thesis (Degree of Master of Science) - Universiti Sultan Zainal Abidin, 2021
|
| 504 |
|
|
|a Includes bibliographical references (leaves 87-97)
|
| 505 |
0 |
|
|a 1. Introduction -- 2. Literature riview -- 3. Research methodology -- 4. Result and discussion -- 5. Conclusion
|
| 520 |
|
|
|a Interaction is central to any learning experience, especially in the online learning
method. Massive Open Online Courses (MOOCs) is one of the learning platforms to
support open leaming education as well as classroom teaching to enhance student
teacher interaction during the teaching and learning process. Studies show that the lack
of interaction between students with the course taken has an impact on the students'
performance i.e. whether they pass, fail or drop the course. Therefore, the objective of
this study is to analyse the significant attributes that may affect their interaction
performance and learning behaviour. This study also proposes a classification model
for the performance of students using MOOCs. For this purpose, the research follows
the Knowledge Discovery in Database (KDD) methodology starting with data
collection, data pre-processing, data analysis by using feature selection method,
classification algorithm, and finally evaluating or interpreting the knowledge gained.
The dataset that has been used in this study is obtained from the Open University
Leaming Analytics Dataset (OULAD). The datasets are tested by applying six
classifiers namely Naive Bayes (NB), Nearest Neighbour (NN), Multi-Class Classifier
(MCC), Rules One R (OR), 148, and Random Tree (RT). The results show that the
proposed classification model developed can classify students according to their
academic achievement and interaction with the MOOCs. It is also found the 148
Classifier is the most appropriate model to classify the performance, 148 Classifier
shows the highest result among the six models that have been applied to the OULAD
data sets with an accuracy of96.8%. It also found that the attributes namely studentfinal
result has a significant relationship with sum click, date submitted, id assessment, imd
band, age band, gender, studied credit, highest education and assessment date which
can contribute to the effectiveness of the students' performance. In conclusion, this
result is vital as the educators can focus on developing the underperforming students
once they have been successfully identified by using such attributes in the MOOCs
platform.
|
| 610 |
0 |
0 |
|a Universiti Sultan Zainal Abidin
|x Dissertation
|v Faculty of Informatics and Computing
|
| 650 |
|
0 |
|a Educational technology
|z Malaysia
|
| 650 |
|
0 |
|a MOOCs (Web-based instruction)
|z Malaysia
|
| 710 |
2 |
|
|a Universiti Sultan Zainal Abidin .
|b Faculty of Informatics and Computing
|
| 999 |
|
|
|a 1000183192
|b Thesis
|c Reference
|e Tembila Campus
|