Quality Education through Data Mining: A Case Study using Educational Performance Data
This study uses data mining to alleviate educational inequity by forecasting student performance and dropout rates. Using a large dataset containing various attributes about student demographics, academic background, and socioeconomic factors, we used a Knowledge Discovery in Databases (KDD) process that included situation understanding, data understanding, preparation, transformation, modeling, interpretation, and proposed actions. The chosen prediction models, such as decision trees and logistic regression, have yielded encouraging results in identifying critical elements influencing educational attainment. This research not only provides significant insights for educational institutions and policymakers, but it also helps to promote equal quality education, which aligns with UN Sustainable Development Goal 4. Version note: Author manuscript prepared as a research paper for INFOSYS 722: Data Mining and Big Data, University of Auckland. Listed as a preprint on ResearchGate on 29 May 2025: https://www.researchgate.net/publication/392166520. This public archive copy removes the author’s student identifier and updates PDF metadata; the scientific text is otherwise unchanged.
Authors
- Ziyi Wang
Institutions
- University of Auckland (NZ)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-15
- DOI
- https://doi.org/10.5281/zenodo.24224155
- Primary Topic
- Online Learning and Analytics
- Type
- preprint