Citation

AlMaqbali, Said (2025) AI Decision Strategy: Machine Learning Identification of Job titles and experience levels based on Skills and Responsibilities. In: 5th International Conference of Science and Information Technology in Smart Administration, ICSINTESA 2025, 29 November 2025, Yogyakarta.

Text IEEE Xplore Full-Text PDF_9.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Successful delivery of engineering projects is aided by proper management of talent. This study uses a quantitative experimental design that focuses on using machine learning (ML) classification models to classify job titles and levels of experience using the skills and responsibility descriptions of the candidate. Through the use of natural language processing (NLP) using Python and supervised learning, we illustrate the ways predictive analytics can assist both HR and project managers in determining the alignment of candidates with the position, as well as the way to compose teams in a manner that is most likely to achieve success. Our models predicted job titles with an accuracy of 75.2% and experience level with 92.5% accuracy and detailed precision, recall and F1-score analysis have been done on a variety of jobs and seniority levels. Word clouds, network graphs, and bar charts are the visualization methods that demonstrate the important skills distribution and interrelations between the roles. The results demonstrate the potential and the existing shortcomings of ML-based decision support in the context of the project management of engineering projects, providing practical recommendations to conduct further research and implement them into practice.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Machine learning
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	20 Apr 2026 03:12
Last Modified:	20 Apr 2026 03:12
URII:	http://shdl.mmu.edu.my/id/eprint/15764

Downloads

Edit (login required)