An Analysis on Academic Performance of Students using a Hybrid Model for Higher Education

Abstract

Abstract-Education has been viewed as a key aspect in contributing to the welfare of the country. In modern era, Educational Institutions strive hard to improve the quality of education they render to the student community. Though many factors determine a good institution, the academic performance of the students pay a vital role in it. Data mining is a technique used to bring out such hidden knowledge which exists in the form of raw data in the repository. Many socio economic, Non-academic and academic factors influence the performance of the students. There are many well -known data mining classification algorithms such as ID3, SimpleCART, J48, NB Tree, MLP, Bayesnet etc which are used to predict the student performance. The model proposed here is mainly focused on finding the prediction accuracy of academic performance of students using a hybrid model which is a combination of two classification algorithms ID3 and MLP. The experimental model also prove that the accuracy can be improved by intelligently generating the training dataset for the hybrid model. Keyword-Educational Data Mining, Academic Performance, Prediction, Classification, Hybrid, ID3, MLP I. INTRODUCTION In this modern era, technological revolution and population explosion have changed the scenario of the Higher Education System. Educational institutions are becoming more competitive because of the number of institutions growing rapidly. To stay afloat, these institutions are focusing more on improving various aspects and one important factor among them is quality learning. For providing quality education, the institutions need to know about their strengths which are explicitly seen and which are hidden. To be competitive, the institutions should identify their own strengths that are substantially hidden and implement a technique to bring them out. In recent years, Educational Data Mining has put on a massive recognition within the research realm for extracting knowledge from a huge dataset. Data mining is widely used on educational dataset and it is termed as Educational Data mining (EDM). EDM has become a very useful research area [1]. Educational Data Mining refers to techniques, tools, and research designed for automatically extracting the pattern from large repositories of data generated by or related to people's learning activities in educational settings. Key uses of EDM include learning and predicting student performance in order to recommend improvements to current educational practice. EDM can be considered as one of the learning sciences, as well as an area of data mining [2]. The technique behind the extraction of the hidden knowledge is a Knowledge Discovery process that extracts the knowledge from available dataset and creates a knowledge base for the benefit of the institution.Many socio economic, non-academic and academic factors influence the performance of the students. Students' data was collected, pre-processed and data mining techniques are applied to discover association, classification, clustering and outlier detection ru...

Modelo Estadístico para determinar los factores académicos en los Resultados de las Pruebas Saber Pro

“…Debido a esto, la minería de datos surge como una posible respuesta, a la necesidad de analizar, manipular y extraer conocimiento de los datos, toda vez que esta hace uso de sofisticados algoritmos que permiten encontrar patrones ocultos en un conjunto de datos y predecir comportamientos; por lo que el uso de la minería de datos para el análisis de información de índole académica, resulta pertinentes según estudios adelantados [2,3,4,5,6].…”