Design and implementation of a webbased university admission and placement neural network model

Project Abstract

<p> </p><p>Every year the number of applicants seeking admission into Nigerian Universities<br>increases by leaps and bounds although the Universities lack the commensurate<br>facilities to meet the challenges of admitting the high number of applicants. For this<br>reason, the admission officers have to manually evaluate every candidate’s data against<br>the set admission requirements to screen the applicants in order to select the number of<br>candidates that their universities can accommodate. The procedures involved are very<br>cumbersome, time consuming and prone to a lot of human errors and irregularities.<br>Many candidates miss out on the admission every year, and the most painful aspect of<br>this manual process is that many who are not qualified for a particular course end up<br>being given placement into such courses while the more qualified ones are left out.<br>Consequently, for lack of aptitude for the course, the students struggle through and<br>many even resort to cheating their way through examinations and then graduate out of<br>the Universities ill-equipped for the job market and the society. On the other hand,<br>some of the less fortunate but qualified ones who are not given University admission<br>year after year, become so frustrated over time and end up in hideous lifestyles.<br>Whichever way, the society suffers and national growth is hindered.<br>In this work, a web-based model was designed to considerably take care of the above<br>problems. The system was developed to provide a time-efficient, detailed and unbiased<br>automated procedure for selecting the most qualified candidates for admission into<br>universities, and ensure that qualified candidates, who fail to meet the requirements for<br>a particular course, are automatically placed into other courses for which they meet the<br>admission requirements and where vacancies exist, using neural network model. The<br>model also provides an avenue for students self-screening admission system.<br>8<br>The system design, implementation and results are presented in chapter four. The<br>implementation was based on AMP (Apache, MySQL, and PHP) open source solutions.</p><p>&nbsp;</p><p>&nbsp;</p> <br><p></p>

Project Overview

<p> </p><p>INTRODUCTION<br>1.1 Background to the Study and Statement of the Problem<br>Higher education in Nigeria can be traced to 1932 when Yaba Higher College<br>was established for the purpose of producing assistants who would relieve the<br>then colonial administrators of menial tasks. Thus in 1940, the University<br>College, Ibadan was established but the programmes offered there and then were<br>narrow because the agenda of the colonial administration did not include the<br>training of high-level manpower for many of the professions. The Ashby<br>Commission in 1960, recommended the establishment of regional universities in<br>the then three regions of Nigeria. Three universities were established: the<br>University of Nigeria, Nsukka (1960) in the Eastern region; the University of<br>Ife, now Obafemi Awolowo University (1961) in the Western region and<br>Ahmadu Bello University, Zaria (1962) in the Northern region, while the<br>existing University College, Ibadan was granted full-fledged University status in<br>1962. Also, the University of Lagos, Akoka came into existence in 1962 and as<br>a city University, it provided courses in law, social sciences, medicine,<br>humanities, engineering and part-time programmes for working students. Lastly,<br>the University of Benin was established in 1970, making the sixth of the<br>Universities that have come to be known as Nigeria’s first generation<br>Universities (Adesina, 1988).<br>Today the higher education system in Nigeria is composed of universities,<br>polytechnics, institutions of technology, colleges of education that form part of,<br>or are affiliated to, universities, and professional, specialized institutions. They<br>can be further categorized as private, state or federal owned institutions. Federal<br>14<br>universities categorized as first, second, or third generation universities, are<br>owned and funded by the federal government, while state universities are owned<br>and financed by the states (there are 36 states in all), and private universities are<br>owned and funded by individuals or religious organisations. As seen from<br>above, the first generation universities are the six universities established in the<br>1960s and early 1970’s; second generation universities are seven universities<br>established in the mid 1970’s; while third generation universities refer to the<br>eleven institutions, including the universities of technology, established in the<br>1980’s and 1990’s (Hartnett, 2000).<br>According to the National Universities Commission (NUC)’s report on the<br>results of the November 2005 System-Wide Accreditation Exercise, there are<br>twenty-five (25) federal universities including three (3) universities of<br>agriculture, twenty (20) state universities, twenty (23) private universities, five<br>(5) degree-awarding colleges of education, sixty-nine (69) National Certificate<br>in Education (NCE) – awarding colleges of education, one (1) military<br>university, four (4) inter-university centres. This gives a total of one hundred<br>and forty-two (142) higher education institutions excluding the polytechnics and<br>the ten (10) newly approved private universities in 2006.<br>Higher education in Nigeria can be further divided into the public or private, and<br>the university or non-university sectors. Public universities, owned by the<br>federal and state governments, dominate the higher education system. The nonuniversity<br>sector is composed of polytechnics, institutions of technology,<br>colleges of education, and professional institutions. There is no sharp distinction<br>between the university and the non-university sectors; most of the institutions in<br>the latter sector are affiliated with universities.<br>15<br>1.1.2 Access to Higher Education in Nigeria<br>There are three levels of university education in Nigeria. The university level<br>first stage offers a Bachelor’s degree after a minimum of three years and a<br>maximum of six years study (e.g. in medicine). The university level second<br>stage offers a Master’s degree following one year of post-Bachelor’s study or<br>one of post-graduate diploma study and a year of post-Bachelor’s study in the<br>relevant discipline. The university level third stage offers doctorate degree of<br>two to three years duration after the Master’s degree. To gain admission into the<br>first level of university education, a potential student has to pass the competitive<br>University Matriculation Examination (UME).<br>In Nigeria and in fact most nations of the world, the University is the highest<br>citadel of learning for the production of high-level human resources for the<br>labour market. In recognition of this and the role of higher education in<br>perpetuating national unity, the Federal Government of Nigeria took appropriate<br>steps to ensure equity with regard to access to university education. The Joint<br>Admissions and Matriculation Board (JAMB) was created by Act No. 2 of 1978<br>of the Federal Military Government (JAMB, 2004). The main aim for the<br>establishment of the Joint Admissions and Matriculation Board (JAMB) was to<br>provide an opportunity for eligible Nigerians to have access to university<br>education, and to diversify the intakes, and achieve a high rate of national spread<br>in the placement of applicants into Nigerian universities (JAMB, 2004). In<br>addition, the JAMB was to place suitably qualified candidates into the existing<br>tertiary institutions after taking into account the vacancies available in each<br>tertiary institution. Placement was to be done on the basis of merit, catchment<br>16<br>area, and with a special focus on females and the Educationally Less<br>Advantaged States (Omoike and Aluede, 2007).<br>The Federal Government controls the universities and other higher education<br>institutions through the following organs: the Federal Ministry of Education; the<br>National Universities Commission, which among other things allocates funds to<br>federal universities and also prescribes the spending formula; and the Committee<br>of Vice Chancellors of Nigerian Universities, which acts as a coordinating body<br>and offers advice to government and universities’ governing councils on matters<br>of general and specific concern to higher education.<br>Individual university administration is the joint responsibility of the university’s<br>Governing Council and the Senate. Although the Governing Council remains<br>the highest policy-making body in the university, an appointed Vice-Chancellor<br>acts as the Chief Executive Officer, coordinating both academic and<br>administrative functions. Within universities and colleges, the institutes and<br>centers are more or less autonomous. The university system polity consists of<br>three distinct categories of staff viz: administrative, academic and technical,<br>each having a union that protects the interest of members. For instance, the<br>Academic Staff Union of Universities (ASUU) safeguards the interests of the<br>academicians in the Nigerian university system. Most universities operate the<br>semester system, where a session comprises of two semesters. The university<br>academic year begins in October and runs through to July.<br>The Federal Government in order to make education relevant to the needs and<br>aspirations of the people and so bring about the desired development reviewed<br>her educational system by introducing the 6-3-3-4 system of education<br>17<br>(Maduewesi and Imhanlahimi, 2006). By this system, before reaching university<br>entrance level, students undertake 12 years of school education, the first six<br>years of which are spent at primary school while the remaining six years are<br>split between junior and senior secondary school education. Then the last four<br>years of the education system are spent on an average four–year course in the<br>university.<br>Generally, students are 18 years old at the start of their university education,<br>though some students are able to gain admission at the younger age of 16.<br>Students may be admitted into the first year of a four-year degree course based<br>on results achieved in the Universities Matriculation Examination (UME)<br>conducted by the Joint Admissions and Matriculation Board (JAMB). Entrance<br>to the first year of a three-year programme is based on results obtained in the<br>West African School Certificate (WASC) O level, in addition to either the<br>General Certificate of Education (GCE) A level, or equivalent examinations<br>such as the Interim Joint Matriculation Board Examination (IJMBE), and the<br>National Diploma (ND) certificates.<br>Admission is through the Joint Admissions and Matriculation Board (JAMB) for<br>both UME and direct entry applicants. For UME, interested applicants purchase<br>the JAMB form and subsequently write the University Matriculation<br>Examination (UME). Based on the score profile of applicants and the<br>recommended enrolment figure for the admission year, a minimum cut-off score<br>is usually proposed for UME admissions. For instance, for the 2005/2006<br>session, a total enrolment figure was proposed as follows: ARTS (2139) and<br>18<br>SCIENCE (3210) for Ahmadu Bello University, and the proposed merit cut-off<br>score was 234 for that session.<br>1.1.3 Statement of the Problem<br>As stated in the preceding section, the Joint Admissions and Matriculations<br>board has the responsibility of placing suitably qualified candidates into<br>Nigerian tertiary institutions after taking into account the vacancies available in<br>each institution. However, the records show that well over 500,000 candidates<br>seek placements into universities annually in Nigeria and only about 13% (on<br>approximation) of them secure admission, which is a far cry from the target.<br>Admission decisions are made by educational institutions by considering a<br>variety of factors. Some of the evaluation criteria normally used are: JAMB<br>UME subject combination; university’s admission requirements; overall scores<br>in JAMB UME results; UME merit cut-off score; five credits obtainable in ‘O’<br>level certificate (in not more than two sittings); catchment area considerations;<br>and educationally less developed states (elds) considerations. In 2006, the Post<br>UME criterion was introduced whereby, a potential candidate having sat and<br>passed the JAMB UME, is examined by his/her University of choice. The Post<br>UME criterion further narrow down or increases the chances of the candidates<br>being selected and placed in the course of their choice or any available course.<br>Faced with all the multiple criteria as stipulated above, the admission officer<br>manually evaluates every candidate’s data against the various admission<br>19<br>requirements before taking admission and placement decisions. This process is<br>quite cumbersome and highly not cost-effective in terms of cost and processing<br>time. Also, as with every manual process, it is fraught with inaccurate decisions<br>resulting from avoidable human processing errors and at times deliberate<br>manipulations to achieve some unwholesome personal aims like admitting<br>unqualified candidates who has personal affiliations with the admission officer.<br>The inherent problems with the manual system of admission gave rise to the<br>need for development of a computerised model that can be used to carry out the<br>admission procedures with a view to:<br>(a) reducing to the barest minimum the admission processing cost and time<br>in terms of man-hours<br>(b) removing all elements of human errors be they intentional or<br>unintentional<br>(c) streamlining admission processing work<br>(d) making admission decisions more objective and impartial<br>(e) admitting only qualified candidates and<br>(f) providing a self-screening and evaluation mechanism for candidates.<br>1.2 Research Motivations<br>Every year more than 500,000 applicants seek placement to universities. Of<br>these applicants only about 13% are selected for admission. That shows that a<br>large number of these candidates miss out on the opportunity of being admitted<br>into the Universities and, for some candidates, this trend continues for many<br>years leading to frustrations.<br>20<br>This problem has remained difficult to surmount because, for any admission and<br>placement process, several admission criteria are considered and the candidates’<br>data would have to be manually evaluated against the various admission<br>requirements by each university’s admission officer before selecting the few<br>qualified candidates. And this is not an easy task.<br>Every Nigerian that seeks admission into the University must have one thing or<br>the other to do with one of the following agencies:<br>i. Joint Admission and Matriculation Board (JAMB);<br>ii. West African Examination Council (WAEC);<br>iii. National Examination Council (NECO);<br>iv. National Teacher’s Institute (NTI);<br>v. Interim Joint Matriculation Board (IJMB).<br>Adebiyi (2006) developed a web-based model for JAMB candidates’ admission<br>and placement into Nigerian Universities. In addition to the general admission<br>requirement of five ‘O’ level credits in not more than two sittings, Adebiyi, in<br>his model, took into consideration other admission requirements such as UME<br>subject combination and total score requirements, and University’s course<br>subject combination requirements as the criteria for admission and placement.<br>His model also presented an opportunity for students self – screening admission<br>system.<br>21<br>The model however, did not take into consideration candidates that get<br>University placement based on variables such as catchment areas and those from<br>educationally less developed states (elds). Since these variables are part of the<br>standard criteria in the University admission and placement process in Nigeria,<br>Adebiyi’s model would deny admission and placement to many candidates who<br>these variables are meant to favour.<br>This study therefore makes an improvement on the model presented by Adebiyi<br>(2006). This is achieved by the development of a web-based model that takes<br>into consideration admission requirements other than those considered by his<br>model. These admission requirements include:<br>a. UME subject combination and total score requirements,<br>b. University’s course subject combination requirements,<br>c. ‘O’ level subject combination and credit pass requirements,<br>d. Post UME requirements,<br>e. Candidates from the University’s catchment area, and<br>f. Candidates from the educationally less developed states.<br>In this era of modern computer technology and information science,<br>sophisticated information systems can be built to make decisions or predictions<br>based on information contained in available past data. Such systems are called<br>learning systems and are currently used for the purpose of classification and<br>prediction. A student data evaluation approach based on neural networks was<br>described in his dissertation (Sheel et. al., 2002). This was used to determine the<br>placement of university students into basic mathematics courses. The existing<br>22<br>approach used the results of a mathematics placement examination given to<br>incoming freshmen to determine course placement, but the study suggested<br>neural networks to be suitable alternatives to such examinations.<br>1.3 Research Objectives<br>The objective of this study is to develop and implement a web-based model for<br>the Admission and Placement of Potential Students into Nigerian Universities.<br>The model evaluates potential students’ data against University admission<br>requirements thereby streamlining and automating the processing work involved<br>in student admission and placement procedures. In addition, the model will<br>provide candidates with an opportunity to perform self-screening and personally<br>evaluate their chances of gaining admission and placement into the Universities<br>of their choice.<br>1.4 Research Methodology<br>A review of related literature is made along with a few related works, and a webbased<br>admission and student placement process model was developed. The<br>technological approach to the implementation was based on open source<br>solutions. The system requirement in terms of software and hardware includes a<br>web server, which is apache extended with support for PHP and MySQL<br>relational database.<br>In recognition of the sensitivity of the data contained in the system,<br>communications over the public network are protected with open-ssl library for<br>data encryption and authentication and role-based authorisation was built into<br>23<br>the system to specify access rights to the database system. System testing was<br>carried out succinctly with real test data which were obtained from the faculty of<br>science of the Ahmadu Bello University, Zaria. Also, as a follow up to the testrunning<br>and subsequent debugging of the system, the system was implemented.<br>At the end of it all, potential students should be able to go on the relevant<br>University’s website and verify whether they qualify for admission and<br>placement into that University. This is self-screening.<br>1.5 Limitations to the Study<br>Some limitations to this study are identified in the area of time-constraint and<br>physical restriction to real data. Due to the volume of work involved in this<br>study, much time is needed to design the model, develop the program code, testrun,<br>debug and fully implement the new system, than is available for the period<br>of this study. Other limitations include the unavailability of constant electric<br>power supply, the erratic nature of Internet connectivity in most parts of the<br>country, and the training and retraining of dedicated staff who will monitor the<br>system.<br>1.6 Contributions to Knowledge<br>By the development of this Web-based Admission and Placement model,<br>potential candidates’ data are evaluated against the university admission<br>requirements and the most qualified candidates are thereby selected. The new<br>system has tremendous contributions to knowledge, as seen below:<br>(i) government’s policy to promote higher education, learning and research<br>is realised since the automated system frees a lot of man-hours to staff<br>24<br>involved in the cumbersome, time-consuming admission process, and,<br>the recovered time can be gainfully employed in capacity building.<br>(ii) many more people now have fair opportunity to increase in knowledge<br>because, when candidates fail to meet some of the admission<br>requirements for their first course, the automated system automatically<br>considers such candidates for placement into any other course for which<br>they meet the requirements, provided there are vacant positions.<br>(ii) a time-efficient, detailed and unbiased automated procedure for selecting<br>the most qualified candidates for admission into universities is put in<br>place and most (if not all) of the bottlenecks that is rife with the current<br>time consuming and error-prone manual admission and placement<br>processes becomes history.<br>(iii) the prevalence of cheating is reduced and hard work is encouraged<br>through the student self-screening admission system .<br>(iv) University admission and placement is made more reliable and this goes<br>to reduce the tendency to corrupt practices even in the larger society.<br>(v) since it provides better admission opportunity for qualified candidates,<br>better qualified graduates will now be turned out into the job market as<br>opposed to the output that comes from persons who struggle through the<br>universities because they were never qualified to be there in the first<br>instance.<br>(vi) with qualified candidates admitted, lecturers have more time on their<br>hands for research which ultimately increases knowledge in all<br>ramifications, since they do not have to over labour themselves to impart<br>25<br>knowledge to students who do not possess the aptitude for university<br>education.<br>1.7. Organization of the Thesis<br>This report is divided into five (5) chapters. In chapter one, a general introduction and<br>background of the study is given, stating the problem, research motivations, objective,<br>methodology, scope of the research and its contributions to knowledge.<br>Chapter two presents a review of various literature in the field of artificial neural<br>networks and some related works that support this research work.<br>In chapter three, the system analysis and modelling is presented showing a detail of the<br>research methodology.<br>The system design and implementation is contained in chapter four with emphasis on<br>the software platform for the implementation, the system hardware requirements and the<br>results of the implementation.<br>Finally, chapter five presents the conclusion and recommendations from the study.<br>26</p><p>&nbsp;</p> <br><p></p>