An Association Rule Mining Based Model to Predict Learning Performance of a student with e-Learning Activity Log Dat
S. Arumugam1, A.Kovalan2, A.E. Narayanan3

1S. Arumugam*, Dept. of Computer Science and Science Applications, Periyar Maniammai Institute of Science and Technology, Thanjavur, India
2A.Kovalan, Dept. of Computer Science, VSB College of Arts and Science, VSB Group of Institutions, Karur, India.
3A.E. Narayanan, Department. of Computer Engineering, Periyar Institute of Science and Technology, Thanjavur, India
Manuscript received on December 14, 2019. | Revised Manuscript received on December 22, 2019. | Manuscript published on January 10, 2020. | PP: 2213-2220 | Volume-9 Issue-3, January 2020. | Retrieval Number: C8970019320/2020©BEIESP | DOI: 10.35940/ijitee.C8970.019320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: There is a wider scope in research on log data of computed aided learning and interactive learning. We have an enormous collection of log data of students’ activities during their learning process. Data Mining(DM) algorithms help us to discover knowledge and information from a huge and complex data sets. In a time-series log data, it is very complicated to verify the DM algorithms to mine the dataset. e-Learning activity log data is taken and converted into categorical data to predict the learning behavior of the students to implement the algorithms. The excavated knowledge can be used to modify the e-learning system. It is very easy from the result to note the slow learners and advanced learners well in advance before conducting an examination. Time series data is a numeric data that measures in a time period in successive order. The dataset used in this work is a UCI EPM dataset. It is a Non-Linear Time-Series Data. The converted dataset is used to apply a rule mining algorithm to predict the performance of a student. The measurements support and confidence will help us to predict the students’ performance. The results also have been compared with other classification mining algorithms. It assists to improve and to build an educational model on e-Learning. In turn it supports students, teachers, and educational system as well Learning Management System. 
Keywords: E-learning, Learning Analytics, Educational Data Mining, Educational Process Mining, Association Rule Mining.
Scope of the Article: Data Mining