The project aimed to present and justify the capability of the data mining process with k-means clustering algorithm in the context of the higher education system.
R (RStudio) | ggplot2 | k-means clustering
Image 1: Data-flow diagram (Level 1)
Image 2: Data (.CSV format) for the 2016 - 2017 academic year (student roll number and percentage)
Image 3: Importing data into R
Image 4: Clusters in scatter plot (2016 - 2017 academic year data)
Image 5: Result from the console
Image 6 and 7: Determining the clusters and optimal value of k using the elbow graph
Image 8 and 9: Finding the cluster's centers for 2016 - 2017 data
• This project was an effort in motivating to advance the traditional educational process with the help of a data mining technique. The presented model can act as a guideline for the higher education system to improve its decision-making processes with the help of generated insights.
• This improvement may bring a lot of advantages to the higher education system by maximizing educational system efficiency, increasing student's promotion rate, retention rate, transition rate, education improvement ratio, success, learning outcome, minimizing the cost of system processes, and decreasing student drop-out rate.