Data Driven Evaluation of the New Learning Paradigm Using Machine Learning for Optimizing Graduate Outcomes
Abstract
In preparing students to face digital literacy and critical thinking transformations rapid, universities are therefore required to design and implement learning processes that are innovative, adaptive, and differentiated, in accordance with the new learning paradigm. However, the unemployment rate in Indonesia remains high approximately 5.98% vocational high school graduates and 4.8% diploma and university graduates. To develop highly skilled human resources, higher education must strengthen the competencies of students as future agents of change entering the workforce. The persistent unemployment rate among diploma and university graduates presents a national challenge that may hinder the progress of human capital development. Therefore, this study aims to develop a classification and association model linking new learning paradigm programs to student learning outcomes, in order to generate new knowledge and identify correlations among grade point average, employment waiting period, occupational field, and graduate income. The research employs a machine learning approach using association rule mining and the K-Nearest Neighbors algorithm to analyze correlations and predict graduate outcomes. Based on data processing of 450 graduate data who participated in the new paradigm learning program, the findings indicate that graduates under the new learning paradigm with grade point average ≥ 3.50 are significantly more likely to secure employment within ≤ 2 months, support = 20%, confidence = 100% based on processing a data set of 450 data. Participants in the teaching assistance program tend to experience longer waiting periods ≥ 6 months and lower initial earnings compared to those from other new learning paradigm pathways. Conversely, graduates involved in certified internships or independent study programs demonstrate higher earnings potential and stronger academic performance. The results confirm that the new learning paradigm exerts a positive influence on graduate employability, income level, and academic achievement, especially through experiential and industry-oriented learning mechanisms.
Article Metrics
Abstract: 36 Viewers PDF: 12 ViewersKeywords
Full Text:
PDFRefbacks
- There are currently no refbacks.

Journal of Applied Data Sciences
| ISSN | : | 2723-6471 (Online) |
| Publisher | : | Bright Publisher |
| Website | : | http://bright-journal.org/JADS |
| : | taqwa@amikompurwokerto.ac.id (principal contact) | |
| support@bright-journal.org (technical issues) |
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0




.png)