Data-driven prediction of career trajectories of industrial engineering students using performance metrics and classification techniques

Authors : Fatma Yerlikaya Özkurt, Ayşe Kuyrukçu

Pages : 404-423

Doi:10.56554/jtom.1745631

View : 37 | Download : 111

Publication Date : 2025-12-30

Article Type : Research Paper

Abstract :The increasing diversity of professions and the multitude of career options have made the process of job selection more challenging and crucial. For aspiring industrial engineers, this choice is particularly complex due to their interdisciplinary education. Their curriculum covers a diverse set of engineering and business courses, including production, modeling, optimization, database, economics, and project management. Unlike some other engineering disciplines, industrial engineering lacks a specific job area definition. This unique situation led to the selection of IE students and graduates as subjects for this study. The study focuses on the mandatory departmental courses for IE students and their corresponding grades. A sample group comprises graduates currently employed in various fields. The primary objective is to establish a relationship between students\\\'coursework and their current job positions through data mining techniques, specifically discriminant analysis and logistic regression. The results, as evaluated by accuracy metrics and classification performance measures, reveal higher rates of correct classification when considering occupational status as the dataset\\\'s response variable. Additionally, discriminant analysis proves effective in categorizing data in relation to industry sectors and occupational status.
Keywords : Kariyer Karar Sürecine Destek, Eğitsel Veri Analizi, Lojistik Regresyon, Diskriminant Analiz

ORIGINAL ARTICLE URL