Building Predictive Models for Data Mining Projects
Allen M. Paz
Allen M. Paz, College of Computing Studies, Information and Communication Technology, Isabela State University, Cabagan, Isabela, Philippines.
Manuscript received on 01 May 2019 | Revised Manuscript received on 15 May 2019 | Manuscript published on 30 May 2019 | PP: 492-498 | Volume-8 Issue-7, May 2019 | Retrieval Number: F3751048619/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper focused on building predictive models for data mining projects and knowledge discovery functionalities. The objectives are 1) data selection and transformation, 2) Generation of a prediction models using classification data mining techniques, 3) Identification of different attributes which affects retention and performance of students and 4) Comparison of accuracy on the classification techniques used in the prediction models. The study used dataset from the students enrolled in the BS Computer Engineering program. Decision tree classifiers such as ID3, J48 and CART were used to build models. Results of the study showed that when the attribute evaluation was conducted using WEKA (Waikato Environment for Knowledge Analysis), the College Entrance Test (CET) got the highest significant value among the identified attributes in predicting the retention and performance of students while J48 got the highest accuracy rating when classifying instances. However, further research on factors or attributes that influence retention and performance of students should be investigated and to include other programs in the University to improve the accuracy of the results of classification.
Keyword: Data Mining, ID3, J48, CART, CET, GWA, HSGPA, SCHLR.
Scope of the Article: Data Mining Methods, Techniques, and Tools.