Open AccessImplementing Classification Techniques in Predicting Incidents in a Higher Education Institution in the Philippines
Author(s) -
Daniel D. Dasig,
Mary Ann B. Taduyo,
Mengvi P. Gatpandan,
Rudolph Val F. Guarin,
Paulino H. Gatpandan
Publication year - 2020
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.e6802.018520
Subject(s) - chaid , computer science , formative assessment , decision tree , minor (academic) , random forest , machine learning , logistic regression , artificial intelligence , quality (philosophy) , decision tree model , tree (set theory) , psychology , mathematics education , mathematical analysis , philosophy , mathematics , epistemology , political science , law
The holistic success of the student in the university heavily relies on the curricula and student development programs. In this milieu, the increasing demand for designing, implementing, monitoring and controlling of major and minor violations of the students' demands formative, reformative, rehabilitative, and restorative remediation programs. This paper presents the implementation of classification technique in predicting incidents, develop a predictive model, and implement the model in a recommender system. The researchers utilized a Descriptive Developmental research design. During the development, business rules, use cases and processes of an HEI were used in developing the recommender system and evaluated using ISO 9126 for Software Quality. The developed predictive model was tested using Classification and Regression (C&R) Tree, C5.0, Quest Tree, Logistic Regression, random tree and Classification technique. On the basis of the findings, the Classification Technique was adopted since it had a higher accuracy rate. The recommender system helped improve employees in incident resolutions, productivity and efficiency, and have provided a significant reduction of students’ major and minor offences based on the classifiers using the CHAID Algorithm. The researchers recommend that further studies and empirical investigation be conducted on the analytical reports, and other data mining techniques may be applied to further improve the system, processes, and student services.