An Interactive Alumni Tracking Dashboard

The aim of this study is to investigate the conversion of graduates to higher degrees and the employability of graduates at the Universiti Sains Malaysia (USM). The study used a combination of data analysis techniques such as continuous pattern analysis, machine learning algorithms, and time series analysis to predict future trends in graduate conversion to higher degrees and employability, using a Power BI Dashboard. The study found that BSc to MSc and MSc to PhD conversion rates are low, suggesting that regional graduate characteristics in addition to academic performance significantly influence conversion to higher degrees. The study used supervised machine learning models, including Decision Tree, Naive Bayes, Support Vector Machine, and Multi-layer Perceptron, to predict graduate conversion. The SVM model performed better than others with 90% accuracy and F1 scores of 29% and 40% for MSc and PhD conversions, respectively. The study also used logistic regression, SVM, and neural networks to predict graduate employability, and the SVM model outperformed the other models with an accuracy of 91% and an F1 score of 82% for the employed class. Time series analysis using the ARIMA model showed a gradual upward trend for conversion to the MSc and a consistently low conversion to PhD over the next five years. Overall, the study's findings have implications for USM and policymakers, as they underscore the need to invest in programs that facilitate graduate conversion to advanced degrees and improve graduate employability. The study also demonstrated the significance of data analytics techniques and machine learning algorithms in gaining insights into student populations and alumni success at educational institutions.