Researcher Ayser Maysara To’mi, a student in the Master’s program in Data Science and Business Analytics, has defended his thesis titled: “A Machine Learning-Based Analysis of the Impact of University Specialization on the Unemployment Rate.”
This research presented two models for classifying data on graduates in general, and data on some specializations in particular, by employment status (employed, unemployed). This was through applying machine learning algorithms to data from the Labor Force Survey carried out by the Palestinian Central Bureau of Statistics.
The researcher created a classification model using five different algorithms - Random Forest, DT, XGBoost, KNN, and AdaBoost for graduate data in general, and three algorithms were applied to classify the data according to university specialization.
The Random Forest (RF) model achieved an excellent accuracy of 95.8% on the test data, the Decision Tree (DT) also provided a high accuracy of 93.4%. It is followed by XGBoost, KNN, and AdaBoost with 91.8%, 88%, and 76.1% respectively for graduates' general power data.
As for the classification of graduates by specialization, the random forest model showed high accuracy in classifying the status of graduates (employed, unemployed) for the four specializations that were chosen (business, management, and law); (Engineering, Manufacturing and Construction), Education, (Health and Wellbeing). The random forest outperformed both the decision tree and KNN, and achieved the highest accuracy results of 94.9%, 94.7%, 94.1%, and 94% for each specialization on row.
The thesis was supervised by Prof. Dr. Ana Fernández Vilas. The committee of examiners included: Prof. Dr. Manuel Fernández-Veiga and Dr. Amani Owda.
