A New Data Mining Model Adopted for Higher ... - Science Direct

Available online at www.sciencedirect.com

ScienceDirect Procedia Computer Science 65 (2015) 836 – 844

International Conference on Communication, Management and Information Technology (ICCMIT 2015)

A New Data Mining Model Adopted for Higher Institutions a b Mohammed I.Al-Twijri , Amin Y. Noaman a Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah, Saudi Arabia b Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah, Saudi Arabia

* Corresponding author. Tel.: +0-000-000-0000 ; fax: +0-000-000-0000 . E-mail address: [email protected]& [email protected]

Abstract In the current times, one of the biggest challenges that educational institutions face in both the developing and developed nations is the explosive growth of educational data and to use this data to improve the quality of managerial decisions. Data mining techniques are considered as an analytical tool can be used to extract meaningful knowledge from these large data sets. The quality of higher education institutions implies providing the services, which most likely meet the needs of students, academic staff, and other participants in the education system. This paper proposes a new Data Mining model to be applied in higher education institutions. The suggested model assists in decision making process in the strategic levels of higher institutions as well as regulates the disciplines of students’ admission. Key Words: Data Models, Data Preprocessing, Higher Education Institution (HEI), Data Mining, Admission System, Educational Data Mining 1.INTRODUCTION For higher education institutions whose goal is to contribute to the improvement of quality of higher education, the success of creation of human capital is the subject of a continuous analysis. Therefore, the prediction of students' success is crucial for higher education institutions, because the quality of teaching process is the ability to meet students' needs. In this sense important data and information are gathered on a regular basis, and they are considered at the appropriate authorities, and standards in order to maintain the quality are set. One of the biggest challenges that higher institutions face today is how to improve the quality of managerial decisions [1-3]. The managerial decision making process becomes more complex as the complexity of educational entities increase. Educational institute seeks more efficient technology to better manage and support decision making procedures or assist them to set new strategies and plan for a better management of the current processes. Business intelligence has tremendous potential to help admissions and Admission management professionals make decisions in areas other than the applicants’ potential success [4-6]. The amount of data stored in educational databases is increasing rapidly. These databases contain hidden information for improvement of students’ performance. Data Mining can help institutions of higher education to make more effective decisions as to improve the quality of instruction and services. Many applications areas such as banking, retail industry and marketing, fraud detection, computer auditing, biomedical and DNA analysis, telecommunications, financial industry have already been advanced through the sturdy techniques of DM [7-11]. Another application domain that can take advantage of DM techniques is higher learning institution. So, the major objective of this paper is to propose a DM model for filtering out students satisfying all eligibility criteria that satisfy the admission methods. Also, the suggested model could predict the best admission methodology to which a student can be applied to the King Abdul-Aziz University in Saudi Arabia.

1877-0509 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of Universal Society for Applied Research doi:10.1016/j.procs.2015.09.037

Mohammed I. Al-Twijri and Amin Y. Noaman / Procedia Computer Science 65 (2015) 836 – 844

837

2. LITERATURE REVIEW Data mining in higher education is a recent research field and this area of research is gaining popularity because of its potentials to educational institutes. o [25] Gave case study of using educational data mining in Moodle course management system. They have described how different data mining techniques can be used in order to improve the course and the students’ learning. All these techniques can be applied separately in a same system or together in a hybrid system. o [26] Have a survey on educational data mining between1995 and 2005. They have compared the Traditional Classroom teaching with the Web based Educational System. Also they have discussed the use of Web Mining techniques in Education systems. o [27] Have a described the use of k-means clustering algorithm to predict student’s learning activities. The information generated after the implementation of data mining technique may be helpful for instructor as well as for students. o [28] Discuss how data mining can help to improve an education system by enabling better understanding of the students. The extra information can help the teachers to manage their classes better and to provide proactive feedback to the students. o [29] Have described the use of data mining techniques to predict the strongly related subject in course curricula. This information can further be used to improve the syllabi of any course in any educational institute. o [30] Describes how data mining techniques can be used to determine The student learning result evaluation system is an essential tool and approach for monitoring and controlling the learning quality. 3. RESEARCH OJECTS Universities are interested in predicting the paths of students, thus identifying which students will join particular faculty and which students will require a large number of debates. For Universities whose goal is to contribute to the improvement of educational quality process, the success of selection of the student best fit path is the subject of a continuous analysis. Therefore, the prediction of students' path is crucial for Universities, because the best fit path that match with the ability of students' needs. All participants in the educational process could benefit by applying data mining on the data from the higher education system (Fig.1). Since data mining represents the computational data process from different perspectives, with the goal of extracting implicit and interesting samples, trends and information from the data, it can greatly help every participant in the educational process in order to improve the understanding of the teaching process, and it centres on discovering, detecting and explaining educational phenomenon’s [23].

Fig. 1 The cycle of Applying Data Mining in Educational Institutions [24] So, with data mining techniques, the cycle is built in educational system which consists of forming hypotheses, testing and training, i.e. its utilization can be directed to the various acts of the educational process in accordance with specific needs:• Of students • Professors and

838

Mohammed I. Al-Twijri and Amin Y. Noaman / Procedia Computer Science 65 (2015) 836 – 844

•

Administration and supporting administration.

Thus, application of data mining in educational systems can be directed to support the specific needs of each of the participants in the educational process. The student is required to recommend additional activities, teaching materials and tasks that would favour and improve his/her learning. Professors would have the feedback, possibilities to classify students into groups based on their need for guidance and monitoring, to find the most made mistakes, find the effective actions, etc [12-15]. Administration and administrative staff will receive the parameters that will improve system performance. In this paper, the proposed model discusses a better methodology for the quality of education process. The model compare among 8 options for admission of students to different faculties. The following are eight admission alternatives that will be evaluated by using the proposed model:KAU Preparatory year GPA to Faculty (P2F) High school GPA to Faculty (H2F) High school + Qiyas (Aptitude Test) to Faculty.((H+Q)2F) KAU Preparatory year GPA+ Qiyas (Aptitude Test) to Faculty.((P+Q)2F) High school GPA + KAU Preparatory year GPA to Faculty ((H+P)2F( High school GPA + KAU Preparatory year GPA+ Qiyas (Aptitude Test) to Faculty ((H+P+Q) 2F) The last two alternatives 5 and 6 include other two options for each alternative as follows:• a) b) • a) b)

If the main courses is KAU preparatory courses the student has to : Passing all preparatory year courses in KAU. Relative rate for college allocation in KAU. If the main courses is depend on HS courses the student has to : Passing all preparatory year courses in KAU. Relative rate for college allocation in KAU.

So, the number of alternatives becomes now eight as shown in table 1. Table1 the matrix of all alternative systems associated with the needed courses

Dataset Alt.

KAU

HS

Qiyas

The needed courses -

P2F

√

-

-

H2F

-

√

-

-

(H+Q)2F

-

√

-

-

-

(P+Q)2F

√

-

√

Passing all preparatory year courses in KAU.

-

Relative rate for college allocation in KAU. Passing all the High School courses. HS Courses Relative rate for college allocation. Passing all preparatory year courses in KAU. Relative rate for college allocation in KAU.

Mohammed I. Al-Twijri and Amin Y. Noaman / Procedia Computer Science 65 (2015) 836 – 844

(H+P)2F

√

√

-

(H+P+Q) 2F

√

√

√

839

1. If the main courses is KAU preparatory courses the student has to : - Passing all preparatory year courses in KAU. - Relative rate for college allocation in KAU. 2. If the main courses is depend on HS courses the student has to : - Passing all the High School courses. - HS Courses Relative rate for college allocation.

The study of college admissions yields requires the insightful domain knowledge of university enrolment strategies, admission process, and decision-making processes. In KAU, The decision makers at the university and presumably in the Ministry of Higher Education often speculate the following questions: Can the enrolled versus not-enrolled admitted applicants be distinguished in terms of academic achievement (such as high school GPA, scores on the Qiyas (Aptitude Test) or Preparatory year Assessment Test. On obtaining answers to this question, what is the methodology and approaches can be adopted to enrol students to the most appropriate colleges? If the suitable of university admission strategies can be identified and predicted, the university will be able to make better- decisions based on data for future recruitment efforts and budgeting purposes. This research proposes a DM model to filter out students satisfying all eligibility criteria.. Further, the model could recommend the best methodology to which a student can apply, satisfying all the criteria of these recommended programs [16-19]. With the help of DM techniques, it is possible to discover the key characteristics from the admission details of students and possibly use those characteristics for student path prediction. Finally, a sample live data of different colleges in KAU is used to illustrate the effectiveness of the proposed model. 4. THE PROPOSED DATA MINING ADMISSION MODEL (DMAM) The various components of the proposed DM admission model consist of five subsystems given as follows [20-22]: a. The DMAM Database Management System; The DMAM Database Management Component stores information which can be further subdivided into that are derived from KAU's traditional data repositories, from external sources such as the Internet, Ministry of Education DB, Qiyas (Aptitude Test) DB. This is will be from the personal insights and experiences of individual users. b. The DMAM Model Management System; The Model Management Component handles representations of events, facts, or situations using various kinds of models such as probabilistic models, queuing models, and econometric model. c. The DMAM Knowledge Base; DMAM Knowledge base component consists of one or more intelligent systems. There are 8 rules that are classified into 6 groups, as shown in Table 2 preparatory year courses in KAU Table 2: knowledge discovery rules and its description

Rules

R1

Sub-

College’s Allocation

Rules

Criteria(CAP)

Description

R1.1

preparatory year courses in KAU

x

R1.2

HS courses

x

Passing all preparatory year courses in KAU. Passing all the High School (HS) courses.

840

Mohammed I. Al-Twijri and Amin Y. Noaman / Procedia Computer Science 65 (2015) 836 – 844

R2.1

Special Mawzona for preparatory year courses in KAU

R2

R2.2

R3.1

Special Mawzona for courses in HS

KAU General Mawzona

R3

R3.2

HS General Mawzona

R4

Nationality

R5

College capacity

R6

Qiyas (Aptitude Test)

KAU preparatory year courses : x Medical colleges: Medicine, Dentist, Pharmacy, student should get B+ in (English, Biology, Chemistry, and Physics) courses and passing personal Interview. x Faculty of Engineering, student should get B in (English, Math, Physics, and Statistics) courses. x Faculty of Computing and Information Technology, student should get B in English, Computer Skills) courses. x Faculty of Applied Medical Sciences, student should get B in (English, Biology, Chemistry, and Physics) courses. High courses : x Medical colleges: Medicine, Dentist, Pharmacy, student should get (mapping B) in (English, Biology, Chemistry, and Physics) courses and passing personal Interview. x Faculty of Engineering, student should get (mapping B) in (English, Math, Physics, and Statistics) courses. x Faculty of Computing and Information Technology, student should get (mapping B) in English, Computer Skills) courses. x Faculty of Applied Medical Sciences, student should get (mapping B) in (English, Biology, Chemistry, and Physics) courses KAU preparatory year courses x Mawzona - Scientific Track. x Mawzona - Literary Track. x Mawzona – administrative Track. x Mawzona - No preparatory year. High courses x x x

Mawzona - Scientific Track. Mawzona - Literary Track. Mawzona – administrative Track.

x

Saudi or Non Saudi

x

Capacity of each college.

x

Aptitude Test degree

d. DMAM User Interface; the user interface subsystem is that component of DMAM system that allows bidirectional communication between the system and its user. It includes not only the hardware and the software but also the factors that deal with ease of use, human-machine interactions.

Mohammed I. Al-Twijri and Amin Y. Noaman / Procedia Computer Science 65 (2015) 836 – 844

e. Users Web-Portal; It is an interactive visualization web module that allows communications between DMAM system and all university decision makers, staff, students and the system, by specifying a query or task, providing information to help focusing the search. 5. THE SIMULATED RESULTS The purpose of this work is to recommend the best continent Admission methodology to which a KAU can apply, satisfying all the norms of these recommended programs. There has been much controversy as to which method of Admission is better, especially for academic achievement of students in the university and which method of Admission should be placed in the university Admission. The following is the analysis of the obtained results, as well as of the prediction model illustrated in the following section. Fig. 2 and Fig.3 show how the DMAM Admission method and KAU Admission method of students to different colleges have 68% with 1242 students of agreement where there are a variation between the two methods exceeded 32%, which represent 596 students So we have to study this difference to see which Admission method is better than the other.

Fig. 2 DMAM Admission Method and KAU Admission Method of Students of the Different Colleges

Fig.3 Students Number at Different Colleges To analyze the 32% which represent the difference of the two methods, the Fig.4 and Fig.5, depict that about 417 students representing 70% of this difference shows that student GPA is decreased with mean and standard deviation of students GPA as -0.566 and 0.488 respectively, While 179 students representing 30% of student GPA is increased with mean and standard deviation of students GPA as 0.392 and 0.316 respectively. By implication, this large standard deviation (0.488) means that the values in the student GPA data set are farther away from the mean, on average, which means that dispersion of student GPA values is high. This indicates that there are students GPA with very low values. Also, the small standard deviation (0.316) means that the values in a student GPA data set are close to the mean of the data set, on average, which means that dispersion of student GPA values is very low. This indicates that there are students GPA with very low values

841

842

Mohammed I. Al-Twijri and Amin Y. Noaman / Procedia Computer Science 65 (2015) 836 – 844

Fig. 4 Differences Percentage Analysis of the two Admission Methods

Fig. 5 Students number of the two Admission Methods According to the above mentioned results, the DMAM proposed Admission method is the best alternative and recommended to be applied in Saudi universities 6. CONCLUDED COMMENTS In this paper some of the main conclusions of the research are summarized, moreover some of future key points are commented. This research article proposed a new model to facilitate the proper vision of selecting the best enrolment method in order to increase the effectiveness the educational process within Saudi universities and at the same time enhancing the student performance. We introduced eight Admission alternatives by using the same technique with different possibilities of application rules in order to show several advantages and disadvantages of the each alternative used as well as to detect the best Admission method among these alternatives. However the comparative study among all the eight alternatives which in fact are 20 alternatives after excluding the odd alternatives showed that, only three alternatives can be considered. Finally, the DMAM proposed Admission method was the best alternative and recommended to be applied in Saudi universities. It is meaningful to emphasise that the DMAM presented in this research, besides using real data, shows that the proposed model is powerful technique .even though the obtain results had been satisfactory, it is necessary to mention that for the future work s it will necessary to test the system with data from other universities in order to know the consistency and convergence in order to obtain better outcomes of this algorithm. Other important aspect is to obtain relation between different Saudi universities college's Admission methodologies. This type of information will eventually represent an additional element to improve recommendations offered by the DMAM system. REFERENCES 1. Abdul Fattah Mashat et al , (2013). Discovery of Association Rules from University Admission System Data, I.J.Modern Education and Computer Science, 4, 1-7 2. Abdul Fattah Mashat et al,( 2012) “A Decision Tree Classification Model for University Admission System , (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 3, No. 10. 3. Amburgey, W. D., Yi, J. C., (2011) Using Business Intelligence in College Admissions: A Strategic Approach,

Mohammed I. Al-Twijri and Amin Y. Noaman / Procedia Computer Science 65 (2015) 836 – 844

International Journal of Business Intelligence Research, 2(1), 1-15, January-March 2011, 4. Arora R. K. and Badal D. (2013). Admission Management through Data Mining using WEKA, International Journal of Advanced Research in Computer Science and Software Engineering, 3(10), 674-678 5. Bhardwaj B. K. and Pal S. (2011) Data Mining: A prediction for performance improvement using classification, International Journal of Computer Science and Information Security, Vol. 9, No. 4. 6. Buldu, A. and Üçgün, K. (2010) Data mining application on students’ data, Procedia Social and Behavioral Sciences 2 (2010) 5251–5259. 7. Bunkar, K., Bunkar, R., Singh, U. K., Pandya, B. (2012) Data Mining: Prediction for Performance Improvement of Graduate Students using Classification, 2012 IEEE 8. Chang, L. (2006) , Applying Data Mining to Predict College Admissions Yield: A Case Study, New Directions for Institutional Research, no. 131, pp. 53-68. 9. Chang, T., (2008) Data Mining: A Magic Technology for College Recruitment, Paper of Overseas Chinese Association for Institutional Research (www.ocair.org). 10. Delavari, N., Amnuaisuk, S., Beikzadeh, M. R. (2008) Data Mining Application in Higher Learning Institutions, Informatics in Education, Vol. 7, No. 1, 31–54. 11. Jiawei Han , Micheline Kamber ,(2006), Data Mining: Concepts and Techniques , Second Edition, The Morgan Kaufmann publisher. 12. Ma, Y., Liu, B., Wong, C. K., Yu, P. S., Lee, S. M. (2000) Targeting the Right Students Using Data Mining, ACM 2000, KDD 2000, 457- 464 13. Monika Goyal1 and Rajan Vohra2, “Applications of Data Mining in Higher Education” , IJCSI International Journal of Computer Science Issues, Vol. 9, Issue 2, No 1, March 2012. 14. Natek S. and Zwilling M. (2014) Student data mining solution–knowledge management system related to higher education institutions, Expert Systems with Applications xxx (2014) xxx–xxx 15. Nguyen, H. (2008) Web-based Admission System for Physical Therapy Graduate Program, Master thesis of Software Engineering, The Department of Computer Science and the Faculty of the University of Wisconsin-La Crosse La Crosse, Wisconsin. 16. Niemi D. and Gitin E. (2012) Using Big Data To Predict Student Dropouts: Technology Affordances For Research, IADIS International Conference on Cognition and Exploratory Learning in Digital Age (CELDA 2012). 17. Ogor, E. N. (2007) Student Academic Performance Monitoring and Evaluation Using Data Mining Techniques, Fourth Congress of Electronics, Robotics and Automotive Mechanics, IEEE Computer Society, 354-359 18. Osmanbegović, E. and Suljić, M. (2012) Data Mining Approach For Predicting Student Performance, Economic Review – Journal of Economics and Business, Vol. X, Issue 1, May 2012, pp. 3-12. 19. Rupnik, R., and Kukar, M. (2007) Data Mining Based Decision Support System to Support Association Rules, Elektrotehniški vestnik 74(4): 195-200, 2007. 20. Tair et. al., "Mining Educational Data to Improve Students’ Performance: A Case Study", International Journal of Information and Communication Technology Research, Volume 2 No. 2, February 2012. 21. Tang H. and McDonald S. (2002) Integrating GIS and Spatial data Mining Technique for Target Marketing of University Courses, Symposium on geospatial Theory, Processing and Applications, Ottawa 2012. 22. Vlahos, G. E., Ferratt, T. W., & Knoepfle, G. (2004). “The use of computer-based information systems by German managers to support decision making”. Journal of Information & Management, 41(6), 763–779. 23. El-Halees, A. (2008), Mining students data to analyze learning behavior: a case study, 24. Romero, C. & Ventura, S. (2007), Educational Data Mining: a Survey from 1995 to 2005, Expert Systems with Applications, Elsevier, pp. 135-146. 25. C. Romero, S. Ventura, E. Garcia (2008), "Data Mining in course management systems: Moodle case study and tutorial", Computers & Education, Vol. 51, No. 1, pp. 368-384 26. C. Romero, S. Ventura (2007), "Educational data Mining: A Survey from 1995 to 2005", Expert Systems with Applications (33), pp. 135-146 27. Shaeela Ayesha, Tasleem Mustafa, Ahsan Raza Sattar, M. Inayat Khan (2010), “Data Mining Model for Higher Education System”, Europen Journal of Scientific Research, Vol.43, No.1, pp.24-29 28. K. H. Rashan, Anushka Peiris (2011), “Data Mining Applications in the Education Sector”, MSIT, Carnegie Mellon

843

844

Mohammed I. Al-Twijri and Amin Y. Noaman / Procedia Computer Science 65 (2015) 836 – 844

University 29. W.M.R. Tissera, R.I. Athauda, H. C. Fernando (2011), “Discovery of Strongly Related Subjects in the Undergraduate Syllabi using Data Mining”, IEEE International Conference on Information Acquisition 30. Sun Hongjie (2010), “Research on Student Learning Result System based on Data Mining”, IJCSNS International Journal of Computer Science and Network Security, Vol.10, No. 4