A REVIEW: DATA MINING TECHNIQUES IN EDUCATION ACADEMIA Nandita Priyadarshini Mitrabinda Ray Department of Computer Science & Department of Computer Science & Engineering, Siksha 'O' Anusandhan Engineering, Siksha 'O' Anusandhan University, Bhubaneswar, Odisha, University, Bhubaneswar, Odisha, INDIA. INDIA. Ph : +91-7205346960 Ph : +91-9861039342 92nandita.priyadarshini@gmail.com mitrabindaray@soauniversity.ac.in ABSTRACT The available resources and infrastructure are not sufficient for a One of the main objectives of Indian educational system is student to being knowledgeable [8]. DM is a process that is helpful for searching the concealed evaluating or enhancing the educational organization. Data information from a large amount of data. It describes the data Mining (DM) is the process of searching the concealed from different origin and it converts into meaningful information. information from a large quantity of data set. It analyzes the data If we will implement massive technology infrastructure for our from different source and it converts into meaningful information. education system then, it must motivate students to go for a better There are a lot of advantages of data mining technique in education.Before utilization of DM process, Pre–processing is education sector. Utilization of DM techniques in education sector most needed method to describe the data sets. Pre-processing is a developing and new growing research area. It is also known methods remove the extraneous information from the collected as Educational Data Mining. The Educational Data Mining is data and keep the relevant information. It converts all the attributes to its category. Figure1 shows the architecture of DM involved with developing the methods that helps to search specific technique :- types of data sets that come from education surroundings. Its main objective is to gets the new learning techniques and upgrade academic result. The use of DM techniques are discussed to increase the performance of the process of higher education system. Various types of classification, clustering and association techniques are used, Which enhance the student performance, their life process management, selection of courses, to measure their reservation rate and allow the fund management of the organization. Keywords Educational Data Mining (EDM), Classification, Clustering, Association. 1. INTRODUCTION The major goal of any academic institution is to bring the quality of education and increase the total work of an institution by Fig.1 Architecture of Data Mining [1] looking at individual works [7]. Education system in India suffers from some serious lacunae and one of the lacunae is in rural India, In this process, data is assembled from different origin and there is no teaching activity on about 50% of the working days in associated the assembled data in a place. Those collected data are the primary schools. termed as data set or target data. Then the target data is processed Copyright ©2017 for the individual papers by the papers’ authors. Copying in advance and converted into the appropriate format. The data permiŠed for private and academic purposes. Œis volume is published and mining methods are tested on the converted data. Finally the result copyrighted by its editors. is presented in forms of tables and graph, and termed as the knowledge. 2. DATA MINING IN EDUCATION There are various types of classification, clustering, association techniques of DM methods that has been used to improving the SECTOR process education sector. Utilization of the DM techniques in education sector is a developing area for research and also it is termed as Educational 2.1 CLASSIFICATION Data Mining (EDM). The EDM involves with developing the methods that are helpful for searching a specific type of data that The classification technique involves learning and classification comes from the academic sectors. The EDM has given the advice of the data. It is the most frequently used DM method, which is for improved decision making process and will increase better used to develop the classes and assign data set to the respective instructions for the organization. There are a lot of advantages of classes [1]. The target data are evaluated by classification DM technique in education sector. Some advantages of DM in algorithm in learning process. In classification method, the test education sectors as follows: data are utilized to evaluate the efficiency of the classification rules [9]. If the efficiency of rules is acceptable, then rules can be DM helps to anticipate the final result of students. utilized to new sets of data. Classification techniques used in It helps to detect student involvement area and education sector such as: determine student’s performance in various fields. Bayesian Classification is used classify the persons into It is used to maintain the records of students in various classes depend on different attributes regarding education sector in a productive way and used to to their educational qualification [10]. classify the organization. Decision Tree is used to predicting the DM operation in education sector is described in figure2 . student’s academic performance. Random Forest is used to predict the change in behavior on student database [11]. 2.2 CLUSTERING Clustering is the most frequently used techniques of DM which is used in different areas like, it retrieve the information from a large database, in bioinformatics, to recognize the patterns and, for image analysis [2]. Clustering methods are tried to find out the approximate solution of a problem. It is an iterative process of discover the knowledge that involves the trial and failure methods. It will be absolutely necessary to change the parameters of the model and data pre-processing to achieve the desire results [3]. This technique also finds the classes and assigns the particular object to a desire class. It helps to record the academic dataset of a student from the database that contains basic student data like name, age, gender, origin, student category academic program, and academic achievements data. Cluster analysis is not different technique but it can be accomplished by various algorithms Clustering techniques used in education sector such as: Partitioning Methods is divided the students in to different sections according to their academic Fig.2 Application of DM in Education Sector [1] performances. Hierarchical clustering is used for extract the In the figure, student gives information to the content. Then the commonly used items from a large database. content forwards the information to student learning data. It creates a new database of the student. Student information system 2.3ASSOCIATION is the existing database which contains all the information detail about the student. The predictive model checks the information The main goal of the association technique is to search the most from student learning data with student information system. If impressive association and interrelationship between a huge data some information is not perfect in both the data base, then the set. Association technique is used to search the most regularly predictive model sends it to the content to rectify it. If the in available data element in a huge data set. Now a day’s by help of information is perfect, then it sends it to the dashboard. Dashboard association rule many corporate companies are increase their is a combination of faculty and administration. It sends the profits. In education sector, it helps the student to searching useful relevant information about student to faculty or administration. If patterns that are helpful in their education, guiding the student some particular information likes permanent address is required to find out the best fit changing model for student learning. by the faculty, then it sends a request to content or it sends [1] directly to student. Again the process continues as follows. 3. BENEFITS OF DATA MINING IN [3] A. Dutt, S. Aghabozrgi, M. A. B. Ismail, and H. Mahroeian, EDUCATION “Clustering Algorithms Applied in Educational Data Mining” , in International Journal of Information and Electronics Engineering, The use of DM techniques for students is to the get better Vol. 5, pp. 105-108, 2015. opportunities for their carrier counseling. The educational data mining can support both the student and the management to [4] M. Goyal and R. Vohra, “Applications of Data Mining in developing their quality of education. DM with student that means Higher Education”, International Journal of Computer Science it contains the related information about the student like name, Issues, Vol. 9, pp. 114-120, 2012 age, gender, course, address etc. It also helps the student in their better development and to enhance better educational process [6]. [5] C. Romero, S. Ventura, “Data mining in education”, WIREs The DM technique can help in improvements of the student Data Mining Knowl Discov, Vol. 3, pp. 12–27, 2013. academic performance. It also improve the web based educational systems [6] P. Gulati, A. Sharma, “Educational Data Mining For Improving Educational Quality”, International Journal of 4. CONCLUSION Computer Science and Information Technology and Security, Vol. 2, pp. 648-650, 2012. Currently, the education system faces a number of issues. To give a solution for those issues, we use different DM techniques with [7] http://startup.nujs.edu/blog/indian-education-system- our education system. It gives a set of methods, which can help what-needs-to-change/ our educational system process to defeat from those problems and increase the quality of education system. It will empower the [8] https://www.linkedin.com/pulse/indian-education- organization in a proper manner that will helpful a student to system-good-bad-arunesh-goyal being knowledgeable and also helps the teachers to provide a quality of education and the management in increasing the [9] S. K. Yadav , S. Pal,” Data Mining: A Prediction for performance of the organization. Performance Improvement of Engineering Students using Classification”, World of Computer Science and Information Technology Journal, Vol. 2, pp. 51-56, 2012. 5. REFERENCES , [10] S. Karthika N. Sairam, “A Naïve Bayesian Classifier [1] V. Kamra, Johina, “A Review: Data Mining Technique Used for Educational Qualification”, Indian Journal of Science and In Education Sector”, International Journal of Computer Science Technology, Vol. 8, pp. 1-5, 2015. and Information Technologies, Vol. 6, pp. 2928-2930, 2015. [11] K. Prasada Rao, M.V.P. Chandra Sekhara Rao, B. Ramesh, [2] P. Veeramuthu, R. Periyasamy, V. Sugasini, “Analysis of “ Predicting Learning Behavior of Students using Classification Student Result Using Clustering Techniques”, International Techniques”, International Journal of Computer Applications, Journal of Computer Science and Information Technologies, Vol. Vol. 139, pp. 15-19, 2016. 5, pp. 5092-5094, 2014.