Skip navigation
st. Mary's University Institutional Repository St. Mary's University Institutional Repository

Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/5273
Full metadata record
DC FieldValueLanguage
dc.contributor.authorGirma, Sisay-
dc.date.accessioned2020-04-07T11:26:08Z-
dc.date.available2020-04-07T11:26:08Z-
dc.date.issued2019-03-
dc.identifier.uri.-
dc.identifier.urihttp://hdl.handle.net/123456789/5273-
dc.description.abstractNowadays student attrition became a universal problem in most higher education. To improve student retention one should understand the non-trivial reason behind student attrition. Student attrition and retention in private higher institution education(PHIE) can be affected by a wide variety of factors, these factors include, demographic, social, economic, academic and institution aspects, are the major contributing aspects that leads to attrition and retention of students in higher education. The main objective of this study is to develop a predictive model using of data mining technology to determine undergraduate students’ attrition or retention in higher. In this study, the hybrid data mining process model is followed. The hybrid data mining process model has six steps such as understanding of the problem, understanding of the data, preparation of the data, data mining, evaluation of the discovered knowledge and use of discovered knowledge. In this study based on the problem understanding, 15 attributes are selected and 7361 instances are used to experiment with designing a predictive model that has a capability of determining students’ status. In this study, the classification algorithms such as decision tree (J48), rule induction (PART and JRIP), and Bayes classifier (naïve Bayes) are used in the model building process. And 10 fold cross-validation and 66% split test option are used to train and test the classifier model. Among the four algorithms tested, decision tree classifier (J48) algorithm scored the highest accuracy of 91.40%followed by PART, JRIP, and naïve Bayes algorithms respectively. Depending on the extracted hidden pattern using J48 algorithm, financial sources (self-sponsored and parent-sponsored, and scholarship), division (regular and extension), types of preparatory attended school (private and public), department (computer science, accounting, marketing management, hotel and tourism, and management), background of study (social and natural), and preparatory completion year, before(1994EC-2001EC) and after (2002EC- 2009EC)) were identified as the major contributing factors behind student attrition and retention(graduated ). The data obtained from SRMIS (student record management information system) was in two table format. So merging the two tables into one table format was the major challenge of this study. It is also difficult to get well organized, correct and quality data for the mining tasks. So we suggest educational institutions to maintain their data symmetrically for data analyses.en_US
dc.language.isoenen_US
dc.publisherSt. Mary's Universityen_US
dc.subjectEducational data miningen_US
dc.subjectattrition, J48 decision treeen_US
dc.titleDeveloping a Predictive Model to Determine Higher Education Students’ Academic Status Using Data Mining Technologyen_US
dc.typeThesisen_US
Appears in Collections:Master of computer science

Files in This Item:
File Description SizeFormat 
Sisay Girma.pdf2.42 MBAdobe PDFView/Open
Show simple item record


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.