Predicting chronic absenteeism using educational data mining methods
| dc.contributor.author | Özdemir Ş. | |
| dc.contributor.author | Çınar F. | |
| dc.contributor.author | Coşkun Küçüközmen C. | |
| dc.contributor.author | Merih K. | |
| dc.date.accessioned | 2024-03-13T10:01:07Z | |
| dc.date.available | 2024-03-13T10:01:07Z | |
| dc.date.issued | 2018 | |
| dc.department | İstanbul Beykent Üniversitesi | en_US |
| dc.description.abstract | The rate of chronic absenteeism is important in assessing the validity of current educational practices conditions. Every student who exhibits this behavior faces the risk of failing to progress to higher level of education and/or dropping out/leaving the school. Students in this risk group represent not only a problem from an educational standpoint but also a potential and multifaceted problem with respect to participation in the economy, the development of a skilled labor force, and the ability to become well integrated into society. In the literature for Turkey, the framework of this problem was constructed using statistical methods, and it is important to analyze this problem in greater depth. The main objective of this study is therefore to employ educational data mining methods to predict cases of chronic absenteeism at high school level. The data, compiled from 2,495 students from different districts of Istanbul, was prepared for data mining operations based on the CRISP-EDM steps. The analysis process was conducted using R language and R language packages due to their flexibility and strength. The study results revealed that the random forest algorithm is able to establish a more successful model, while the C4.5 algorithm more accurately describes the problem in terms of decision rules. © 2018, Springer International Publishing AG, part of Springer Nature. | en_US |
| dc.description.sponsorship | This paper was supported by ‘Human Development Research Award’ given by Professor Çiğdem Kağıtçıbaşı at UNESCO Chair on Gender Equality and Sustainable Development, Koc University | en_US |
| dc.identifier.doi | 10.1007/978-3-319-64554-4_36 | |
| dc.identifier.endpage | 526 | en_US |
| dc.identifier.issn | 2213-8684 | |
| dc.identifier.scopus | 2-s2.0-85059097061 | en_US |
| dc.identifier.scopusquality | N/A | en_US |
| dc.identifier.startpage | 511 | en_US |
| dc.identifier.uri | https://doi.org/10.1007/978-3-319-64554-4_36 | |
| dc.identifier.uri | https://hdl.handle.net/20.500.12662/3002 | |
| dc.indekslendigikaynak | Scopus | en_US |
| dc.language.iso | en | en_US |
| dc.publisher | Springer | en_US |
| dc.relation.ispartof | Springer Proceedings in Complexity | en_US |
| dc.relation.publicationcategory | Kitap Bölümü - Uluslararası | en_US |
| dc.rights | info:eu-repo/semantics/closedAccess | en_US |
| dc.subject | Chronic absenteeism | en_US |
| dc.subject | CRISP-EDM (cross-industry standard process for educational data mining) | en_US |
| dc.subject | Educational data mining | en_US |
| dc.subject | Machine learning | en_US |
| dc.subject | R | en_US |
| dc.title | Predicting chronic absenteeism using educational data mining methods | en_US |
| dc.type | Book Chapter | en_US |
