Design And Implementation Of A System For Predicting Student Performance Using Artificial Neural Network

Design And Implementation Of A System For Predicting Student Performance Using Artificial Neural Network

Abstract

Artificial intelligence has enabled the development of further sophisticated and more effective pupil models which represent and descry a broader range of pupil geste than was preliminarily possible. In this exploration, the perpetration of a stoner-friendly software tool for prognosticating the scholars ’ performance by course which is grounded on a neural network classifier will be made. This tool has a simple interface and can be used by an preceptor for classifying scholars and distinguishing scholars with low achievements or weak scholars who are likely to have low achievements. The observed poor quality of graduates of scholars of this institution in recent times has been incompletely traced to crunches of some or utmost of the speaker in the University which goes down to capability to handle the scholars. In this study an Artificial Neural Network( ANN) model, for prognosticating the likely performance of pupil will be developed and tested. The system will be developed and trained using data gauging five generations of graduates from one of the department in the academy. The use of artificial intelligence has enabled the development of further sophisticated and more effective pupil models which represent and descry a broader range of pupil geste than was preliminarily possible.

Chapter One

Preface

During the last many times, the operation of artificial intelligence in education has grown exponentially, prodded by the fact that it allows us to discover new, intriguing and useful knowledge about scholars. Educational data mining( EDM) is an arising discipline, concerned with developing styles for exploring the unique types of data that come from educational environment. While traditional database queries can only answer questions similar as ” find the scholars who failed the examinations ”, data mining can give answers to further abstract questions like ” find the scholars who’ll conceivably succeed the examinations ”. One of the crucial areas of the operation of EDM is the development of pupil models that would prognosticate pupil characteristics or performances in their educational institutions. Hence, experimenters have begun to probe colorful data mining styles to help preceptors to estimate and ameliorate the structure of their course environment.

The main ideal of the admission system is to determine campaigners who would probably do well in the university or can perform well within the academic time or to produce scholars of high grade and intelligence. The quality of campaigners admitted into any advanced institution affects the position of exploration and training within the institution, and by extension, has an overall effect on the development of the country itself, as these campaigners ultimately come crucial players in the affairs of the country in all sectors of the frugality.

lately, still, there has been a conspicuous slide in the quality of graduates of some Nigerian universities. The crunches of the present university admission system, among other factors, have been criticized for this decline. Due to the adding gap between the figures scholars seeking admission and the total available admission places, there has been a corresponding increased pressure on the process. This pressure has lead to rampant cases of admission fraud and affiliated problems.

In Nigeria, scholars are needed to enter secondary academy after spending a minimum of six times of Primary Education and passing a specified National Common Entrance Examination. A pupil also spends a minimal period of six times in Secondary academy at the end of which he or she takes the General Certificate of Education Examination( GCE), also known as the elderly Secondary Certificate Examination( SSCE) or the Ordinary position Examinations. A outside of nine and a minimum of seven subjects are registered for in the examination with Mathematics and English Language being mandatory. Nine possible grades are accessible for each subject; these are A1, A2, A3( distinctions grades) C4, C5, C6,( credit grades), P7, P8( pass grades), and F9( Failure).

Hence this study takes an engineering approach to diving the problem of admissions by seeking ways to make the process more effective and effective. Specifically the study seeks to explore the possibility of using an Artificial Neural Network model to prognosticate the performance of a pupil before admitting the pupil.

Intimately one expects the performance of a pupil to be a function of some number of factors( parameters) relating to the background and intelligence of said pupil. It’s still egregious that it’ll be relatively delicate chancing an logical( or a fine) model that may nicely model this performance/ factors relationship. still one practical approach for prognosticating the performance of a pupil may be by ‘ reasoning ’ from literal data of once scholars ’ background and their associated performances.

The debit then’s the difficulty of opting an applicable function able of landing all forms of data connections as well as automatically modifying affair in case of fresh information, because the performance of a seeker is told by a number of factors, and this influence/ relationship isn’t likely going to be any simple given retrogression model.

An artificial neural network, which imitates the mortal brain in problem working, is a more general approach that can handle this type of problem. Hence, our attempt to make an adaptive system similar as the Artificial Neural Network to prognosticate the performance of a seeker grounded on the effect of these factors.

The results of this vaticination can also be used by preceptors to specify the most suitable tutoring conduct for each group of scholars, and give them with farther backing acclimatized to their requirements. In addition, the vaticination results may help scholars develop a good understanding of how well or how inadequately they would perform, and also develop a suitable literacy strategy. Accurate vaticination of pupil achievement is one way to enhance the quality of education and give better educational services( Romero and Ventura, 2007). Different approaches have been applied to prognosticating pupil academic performance, including traditional fine models and ultramodern data mining ways. In these approaches, a set of fine formulas was used to describe the quantitative connections between labors and inputs( i.e., predictor variables). The vaticination is accurate if the error between the prognosticated and factual values is within a small range

Background Of The Exploration

In machine literacy and cognitive wisdom, artificial neural networks( ANNs) are a family of statistical literacy models inspired by natural neural networks( the central nervous systems of creatures, in particular the brain) and are used to estimate or compare functions that can depend on a large number of inputs and are generally unknown. Artificial neural networks are generally presented as systems of connected “ neurons ” which exchange dispatches between each other. The connections have numeric weights that can be tuned grounded on experience, making neural nets adaptive to inputs and able of literacy. For illustration, a neural network for handwriting recognition is defined by a set of input neurons which may be actuated by the pixels of an input image. After being ladened and converted by a function( determined by the network’s developer), the activations of these neurons are also passed on to other neurons. This process is repeated until eventually, an affair neuron is actuated. This determines which character was read.

The artificial neural network( ANN), a soft computing fashion, has been successfully applied in different fields of wisdom, similar as pattern recognition, fault opinion, soothsaying and vaticination. still, as far as we’re apprehensive, not important exploration on prognosticating pupil academic performance takes advantage of artificial neural network. Kanakana and Olanrewaju( 2001) employed a multilayer perception neural network to prognosticate pupil performance. They used the average point scores of grade 12 scholars as inputs and the first time council results as affair. The exploration showed that an artificial neural network grounded model is suitable to prognosticate pupil performance in the first semester with high delicacy. A multiple feed-forward neural network was proposed to prognosticate the students†™ final achievement and to classify them into two groups. In their work, a pupil achievement vaticination system was applied to a 10- week course. The results showed that accurate vaticination is possible at an early stage, and more specifically at the third week of the 10- week course.

Advising scholars on their class performance and motivating them in order to ameliorate on their performance is an integral part of every instruction. The mechanisms to achieve the below end needed a fashion able of directly prognosticating pupil achievement as early as possible and cluster them for better academic backing. According to Lykourentzou et al,( 2009), pupil- achievement vaticination can help identify the weak learners and duly help them to manage with their academic pursuit. Several styles and systems have been developed for the below task, utmost of which are artificial intelligence- grounded. For case, Lykourentzou etal.,( 2009) estimated the final grades of scholars ine-learning courses with multiple feed-forward neural networks using multiple- choice test data of scholars of National Technical University of Athens, Greece as input. The results attained shows that ANN is91.2 effective. Junemann, Lagos, and Arriagada( 2007) used neural networks to prognosticate unborn pupil training performance grounded on scholars ’ family, social, and wealth characteristics. The forenamed work concentrated on prognosticating the achievement of 15- time-old secondary scholars on reading, mathematics and wisdom subjects in Berlin.

In the Nigeria environment, Oladokun, Adebanjo & Charles- Owaba( 2008) applied multilayer perceptron neural network for prognosticating the likely performance of campaigners being considered for admission into Engineering Course of the University of Ibadan using colorful impacting factors similar as ordinary position subjects ’ scores, matriculation test scores, age on admission, maternal backgroundetc., as input variables. The results showed that ANN model is suitable to rightly prognosticate the performance of further than 70 of prospective scholars.

still, Abass etal.,( 2011) applied another fashion of Artificial Intelligence( AI) i.e., case- base logic( CBR) to prognosticate pupil academic performance grounded on the former datasets using 20 scholars in the Department of Computer Science, TASUED as the study sphere. The high correlation measure observed between the factual graduating CGPA and the CBR prognosticated bones also justify the utility and effectiveness of AI ways in this type of task.

In this exploration work, Artificial Neural Network is used to estimate scholars ’ final grade in the university with a vaticination position of 92.

Statement Of Research Problem

Looking into the institution this days, you’ll discover that 48 of the pupil are actually performing veritably low on their academic position, whom if asked to defend his admission status can not( i.e. sitting for the station test), when proper disquisition is carried out, findings shows that utmost of them have their way into the academy through fix or the so called upper hand. Also another issue or problem for this exploration work is that some of the applied campaigners, some are actually sound and able of performing well when admitted, but because of some factors at the moment or girding the pupil, help the pupil from carrying or securing his admission into the academy. With this little problem I seek to develop a neural network system an artificial bone that will break this problem. Coupled with the stress gone through during the admission and detention in the process that ends up not being done perfect to the standard needed.

The observed poor academic performance of some Nigerian scholars( tertiary and secondary) in recent times has been incompletely traced to crunches of the National University Admission Examination System. It has come egregious that the present process isn’t acceptable for opting potentially good scholars. Hence there’s the need to ameliorate on the complication of the entire system in order to save the high integrity and quality. It should be noted that this feeling of uneasiness of stakeholders about the traditional admission system, which isn’t peculiar to Nigeria, has been an age long and global problem. Kenneth Mellamby( 1956) observed that universities worldwide aren’t really satisfied by the styles used for opting undergraduates. While admission processes in numerous advanced countries has served from, and has been enhanced by, colorful advances in information wisdom and technology, the Nigerian system has yet to take full advantage of these new tools and technology. Hence this study takes an scientific approach to diving the problem of admissions by seeking ways to make the process more effective and effective. Specifically the study seeks to explore the possibility of using an Artificial Neural Network model to prognosticate the performance of a pupil before admitting the pupil.

Objects Of The Study

The primary end of my exploration work is to develop an artificial neural network system that will be able of prognosticating pupil performance.

Some other objects which I’ll be covering in this exploration work are as follows

A system that will enhance the admission process of this institution in terms of admitting the right pupil into the institution.

An easy and friendly stoner interface ANNs which will allow fast operation and analysis of pupil affair.

To determine some suitable factors that affect a pupil’s performance.

To model an artificial neural network that can be used to prognosticate a seeker’s performance grounded on given pre demand data given to it.