CHAPTER ONE

     INTRODUCTION

1.1    BACKGROUND OF THE STUDY

Evaluation of student academic performance is called grading. Academic performance is normally measured by cumulative grade point average (CGPA) attained during a given academic year. Prediction of students’ performance based on their previous academic performances, putting into consideration the CGPA, is a common traditional practice in higher institutions of learning. Several students have low grades while they are in their first year; findings revealed that, stress during this period is associated with overall academic performance with low CGPA inclusive. But the recent need to combine other factors to enhance the prediction of students’ performances has been recognised and the rate of implementation of such is on the increase in many academic institutions. The prediction of students’ performance cannot be over-emphasised as it is a useful tool for policy planners, teachers and students, as this can help to have improved understanding of student’s weakness and bring about enhancement. First year students performance, previous academic performance, and demographic data, student background and demographics, prior educational performance and level, psychosocial factors and approaches to study, and institutional and course factors (Scott & Smart, 2005). Wimshurst & Allard(2008) said students need some form of monitoring especially as regards to their academic performances. A plenty of variables have been found to impact maintenance and execution, falling into various general classes: including sex, identity components, insights and inclination tests, scholarly. Notwithstanding, components discovered to be prescient in a few studies are not generally prescient in others (Deberard, Spielmans and Julka, 2004), due to a limited extent to the routes in which distinctive studies are planned. Undoubtedly, even in the same study with the same approach, results for understudy companions examined at distinctive colleges have contrasted (Vandamme, Meskens and Superby, 2007) and all in all the consequences of specific studies can't be summed up to different situations (Scott and Smart, 2005).

Support Vector Machines (SVMs) also called Support Vector Networks are discriminative classifiers formally defined by separating hyperplane. Support Vector Machines are supervised learning models that possess learning algorithms that analyse data and recognize patterns, used for classification and regression analysis. Originally, the Support Vector Machine algorithm was invented by Vladmir N. Vapnik and the current standard incarnation was proposed by Corrina Cortes and Vapnik in 1993 and published in 1995.

The application of SVM in students’ performance prediction provides efficient and feasible solutions as the knowledge of Support Vector Machine is suitable when prediction is needed. SVM technique is one of the most important techniques to handle prediction and uncertainty bearing in mind evaluating the past academic performance of students in order to establish the risks of failure involves dealing with imprecise data.

However, the major limitations in the current academic prediction include the lack of information behind the prediction method used and the criteria considered in classifying the student’s grade. There are many factors, which were stated earlier, that can be considered as the parameters for student’s performance prediction. It has been observed that there are factors that pose barriers to students and puts them at risk of not attaining and maintaining high grades. Study suggests that there is a need to combine other factors to enhance the Prediction of students in academic institutions.

In this study, evaluating a student’s academic performance is based on the cumulative grade point average (CGPA) obtained at the end of the first year. Universities use cumulative grade point average (CGPA), an example of score aggregation based measure, as a major criterion for student selection. In most Nigerian universities, a CGPA of 3.5 and above is considered an indicator of good academic performance; hence, CGPA is the most common factor used by the academic planners to evaluate progress in an academic environment. At the end of the students’ programs, they will be awarded with specific degrees in appropriate categories or classes. In higher institutions of learning, grading the class performances is a way to classify students according to a pre-defined specification. For instance, student who obtained CGPA of at least 4.50 is classified as a first class student while, student who obtained CGPA between 2.40 to 4.49 points is classified as a second class student.  Thus, student’s CGPA falls into a certain class of categorization depending on his/her CGPA performance. The process of student prediction is based on a procedure where students are classified as either excellent students or regular students. The decision to determine student’s grade is carried out using only one factor, namely the academic performance which is usually derived from the students CGPA as obtained. In this particular scheme, other factors and attributes are not taken into consideration in the student’s classification judgments. The awarded degrees do not indicate the additional skills or the student’s actual capability. Moreover, academic learning may also include other form of learning experiences. Despite its limitations in providing a comprehensive view of the state of students’ performance and simultaneously discovering important details from their continuous performance assessments, it remains the most common factor used by the academic planners to evaluate progression in an academic environment. Furthermore, average score may lead to wrong conclusion. Especially, when details of data from which it is computed are not given. For example, consider a scenario where two students score 50, 60, 70, and 70, 60, 50 in three tests respectively. The average mark obtained by each is 60. Can we conclude, from the average, that intelligence level of both students is the same? Of course not! The data indicates that one student is improving while the other is deteriorating consistently. It may imply that one student is learning consistently from his experience.  Despite these limitations, the CGPA is almost the only instrument that is being used for evaluation by academic planners to Evaluate (Sansgiry, S.S., et al., 2010).  This research investigates the implementation of Support Vector Machine in student’s Performance Prediction.

1.2   STATEMENT OF THE PROBLEM

            There are factors that hinder the academic performance of students. These factors tend to have long lasting effect on the student's academic performance through out his/her time as a student. The factors vary with different individuals. These factors can be addressed to reduce the woeful academic performance of students.

            This project is meant to analyse, demonstrate and predict the performance of a particular student’s with respect to the factors; gender, personality factors, academic, e.t.c., using the Support Vector Machine technique.

1.3  AIM AND OBJECTIVES OF THE STUDY

The aim of this project is to predict a student’s performance using the Support Vector Machines method,

The three objectives of the proposed research are as follows:

·         Identify and select appropriate data mining techniques for developing predictive models.

·         Identify and select appropriate predictor variables/independent variables that can be used as the inputs of predictive models.

·         Validate the developed model using the data collected to identify academically-at-risk students.

1.4  SCOPE AND LIMITATION OF THE STUDY

The study is concerned with developing a model or framework of predictive system by identifying the most suitable factors for predicting first year student performance.

Student academic performance is affected by numerous factors. The scope of the research is limited to the investigation of the effects of a student’s prior achievement, domain-specific prior knowledge, and learning progression on their academic performance. The scope of this project is to predict performances with respect to these factors that influence them. It takes extra effort and time for non-experts to understand the model and utilize it

1.5  SIGNIFICANCE OF THE STUDY

This study makes it easy for the tutor to identify where the student has problem and how to help that particular student thereby improving the student’s grades and overall success in his/her academics. The study helps the institution attain their primary aim of producing academically successful students.

You either get what you want or your money back. T&C Apply

Talk to us right now: (+234)906-451-7926 (Call/WhatsApp)

---