University of Twente Student Theses
Technology-Driven Career Path Prediction : A Case of Southeast Asia's Public Organization
Pasaribu, Juwita Pebriana (2024) Technology-Driven Career Path Prediction : A Case of Southeast Asia's Public Organization.
|
PDF
2MB |
| Abstract: | In recent years, there has been a growing interest in using machine learning to improve human resource management, such as predicting employee attrition, recruitment needs, and performance. This thesis focuses on predicting career paths for employees in government organizations in Southeast Asia, an area previously overlooked by researchers. The goal is to develop a machine learning-based career path prediction framework for civil servants in this region. To achieve this, the research involves a literature review and gap analysis to identify research opportunities and define system requirements. It also includes implementing and analyzing the most suitable machine learning methods for accurate career path prediction. The thesis adopts an empirical research method. Given the complexity and importance of career progression in the public sector, we evaluate four machine learning algorithms to identify the most effective model for accurate career path prediction. The models assessed include Decision Tree (DT), Random Forest (RF), XGBoost, and Multilayer Perceptron (MLP). These models were chosen based on a comprehensive literature review that highlighted their effectiveness in similar applications. Adhering to the CRISP-DM methodology, the research encompasses phases of business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The dataset, sourced from the National Civil Service Agency (NCSA), underwent extensive preprocessing. The evaluation focused on key performance metrics: accuracy, precision, recall, and F1-score. The findings revealed that XGBoost outperformed other models for functional positions, demonstrating superior accuracy and handling of complex data interactions. For managerial positions, while Random Forest exhibited the highest accuracy, MLP showed better precision and F1-score, indicating its effectiveness in minimizing false positives. The work in this thesis significantly contributes to both research and practice by addressing a critical gap in career path prediction within the public sector, specifically for civil servants in Southeast Asia. By developing tailored predictive models based on real-world data, this research provides valuable insights for optimizing workforce planning, enhancing talent management, and informing policy decisions in public service organizations. Moreover, it enriches the global conversation on career path prediction by integrating region-specific data and methodologies, presenting a validated framework that underscores the technical viability and practical advantages of incorporating machine learning into human resource management systems. |
| Item Type: | Essay (Master) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science MSc (60300) |
| Link to this item: | https://purl.utwente.nl/essays/103407 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page