Education sector in India is a growing field that plays a pivotal role in improving the living status. The economic status or the rise of a country depends on the improved education system. According to statistical survey, India after Independence gave more importance to primary education and expanded literacy rate to two thirds of its population. There are several efforts made by the government to improve the literacy rate in India. Despite the education’s sector growth, 25% of its population are still illiterate and the number of enrolment of students to higher education is still in decline. Data mining deals with the process in which we identify and extract all the hidden information from data bases. Educational data mining plays a very important role in identifying, analyzing and visualizing the data to predict students’ performance, their academic achievements, providing feedback for supporting instructors and so on. There are so many factors that affect students’ enrolment to post secondary education. So, the main aim of this research is to identify those factors using data mining techniques which will help the educational institutions, academic heads and also the policy makers of the government schools to take necessary action.
Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
3. INTRODUCTION
A.DATA MINING:
Data mining [6] [7] is the emerging field of applying statistical and artificial intelligence techniques to the problem of finding novel, useful, and non-trivial patterns from large databases. Data Mining is often defined as finding hidden information in a database[8]. Data mining provides many tasks that could help to study the students’ performance[9]. Different data mining techniques are used in various fields of life such as medicine, statistical analysis, engineering, education, banking, marketing, sale, etc (MacLennan. 2005).
B.EDUCATIONAL DATA MINING (EDM)
“Educational Data Mining is an emerging discipline, concerned with developing methods for exploring the unique types of data that come from educational settings, and using those methods to better understand students, and the settings which they learn in.”[1]. Day by day the growth of the data is very rapid and that data need to transformed and converted into an useful information [2]. Educational data mining (EDM) tends to focus on new tools and techniques for discovering patterns in the data. It also gains popularity in the new research areas in higher education. Recent research findings in educational data mining helps the students, institutions and government for improving the quality of education. Inspite of the rapid growth in the education sector , 25% of its population is still illiterate , 15% of the students reach high school, and only 7% graduate[3]. Statistics says according to the year 2011,out of 74% of the literacy rate, only 47% have attained the diploma and post diploma courses[4].Post secondary education plays a vital role in country’s development. But the statistical data proves still major population in India are school dropouts. There are so many factors which affect the students’ enrolment to post secondary education such as family background, school infrastructure and facilities and their psychological behaviours and so on. The main aim of this paper is to identify the reasons for poor enrolment to post secondary education and the result will help the students, management and policy makers to give a better solution. Data mining techniques particularly classification helps to analyze the input data and to develop a model describing important data classes or to predict future data trends.
4. LITERATURE SURVEY
In[11], the author uses the data mining processes, particularly classification to help in enhancing the quality of the higher educational system by evaluating student data to study the main attributes that may affect the students performance in courses. Ayesha et.al [12] used clustering techniques in data mining to analyze students learning behaviour which helped the teachers to identify the drop out ratio to a significant level and improve the performance of the students. Liu Kan [13] designed a course management system on the basis of data mining methods such as classification, association rules and clustering. In [14], the author used different classification algorithms to get useful information to decision-making out of customers’ transaction behaviours. In [15], the author applies four different classification methods for classifying students based on their final grade obtained in their courses. Dr. Surabh paul[16], in his research used classification to evaluate previous year’s student dropout data using Bayesian classification method.
5. STATEMENT OF THE PROBLEM
This minor research aims to study the socio-political factors affecting the students’ enrolment to post secondary education using data mining techniques. These attributes consist of 1)personal information such as age, gender, occupation of the parents, family income, highest educational qualification of the parents, stay, family size.2)institution related information such as type of learning, usage of teaching aids, exposure to ICT, faculty qualification etc 3)psychological information such as social status, illness, disability etc are considered. These attributes were used to predict the students’ enrolment to post secondary education.
6. CONCEPTUAL AND THEORETICAL FRAMEWORK
To build the classification, CRISP methodology is adopted. The proposed methodology is to build the classification model that tests the factors which affect the students’ enrolment to post secondary education.
DATA MINING PROCESS
Knowing the reasons for not continuing their post secondary education can help the teachers and administrators to take necessary actions so that enrolment rate can be improved. Predicting the reason for students not enrolling to post secondary education needs a lot of parameters to be considered. Prediction models that include all personal, social, psychological and other environmental variables are necessitated for the effective prediction and decisions to be made.
A.BUSINESS UNDERSTANDING
Business understanding focuses on the understanding of the project objective and requirements from business perspective then converting it into a data mining problem definition and a plan is designed to accomplish those objectives.
B.DATA UNDERSTANDING
Data set is to get familiar with the data and to identify the problem to discover useful information out of it. Data understanding also helps to examine the quality of data in addressing the questions “Is the data complete? or any missing values?”. The data set used in this study was obtained from the Gottigere Government High School, Karnataka. Initially size of the data is 110.
C.DATA PREPARATION
Data Preparation takes usually 90% of the time to collect, assess, clean and select the data required to construct, integrate and format the data. Identify data sources based on the data available to solve an identified business problem or objective. From the selected data sources, the actual data to be used must be determined [20].
D.BUILDING THE CLASSIFICATION MODEL
The collected attributes may have some irrelevant attributes that may degrade the performance of the classification model; a feature selection approach is used to select the most appropriate set of features. Classification techniques are supervised learning techniques that classify data item into predefined class label [19]. This technique in data mining is very useful from a data set to build the classification model that is used to predict future data trends. With classification, the generated model will be able to predict a class for given data depending on previously learned information from historical data. To explore knowledge discovery decision tree to produce a model with rules in human readable way. The tree has the advantages of easy interpretation and understanding for decision makers to compare with their domain knowledge for validation and justify their decision [19]. Some of decision tree classifiers are C4.5/C5.0/J4.8,ID3 and others.
Generating the Classification rule by applying ID3 algorithm
The classifier identified to implement this model is ID3 algorithm. The decision tree building algorithm ID3 determines the classification of objects by testing the values of the their attributes. It builds the tree in a top down fashion, starting from a set of objects and a specification of properties. At each node of the tree, a property is tested and the results are used to partition the object set. This process is recursively done till the set in a given sub tree is homogeneous with respect to the classification criteria – in other words it contains objects belonging to the same category. This process then becomes a leaf node. At each node, the property to test is chosen based on information theoretic criteria that seek to maximize information gain and minimize entropy. In simpler terms, that property is tested which divides the candidate set in the most homogeneous subsets[17]. For this purpose the WEKA toolkit is used and the attributes are ranked and then the ranked attributes are eliminated by the feature selection approach.
Find Out How UKEssays.com Can Help You!
Our academic experts are ready and waiting to assist with any writing project you may have. From simple essay plans, through to full dissertations, you can guarantee we have a service perfectly matched to your needs.
View our academic writing services
E. EVALUATION:
Evaluation is to check whether we correctly built the model and determines how to proceed and whether to finish the project and move on to deployment phase. Evaluating the results assess the degree to which the model meets the business objectives and also unveils additional challenges, information or hints for future directions. Choosing the proper data mining method is a critical and difficult task in KDD process. To implement this model WEKA Toolkit is used which has a collection of machine learning algorithms for solving data mining problems implemented in Java. Weka has tools for data processing, classification, regression and association, clustering and visualization. It is an open source toolkit for machine learning.
F.DEPLOYMENT:
Deployment phase is to determine how the evaluated results need to be utilized. The knowledge gained has to be organized and presented in the way it is applicable to the end user. This phase may be a final and comprehensive presentation of the data mining results. This CRISP provides a uniform framework for experimenting, analyzing, evaluating and predicting the result
7. SPECIFIC OBJECTIVES:
There are few objectives stated below:
1. This project is a preliminary attempt to help supporting the decision makers of the institution to improve their teaching methodology, and teaching aids and all other infrastructure facilities that they lack.
2. The result evaluated out of this project will motivate the parents of BPL (Below poverty line) towards the values of post secondary education.
3. This project will help the policy makers of our Indian government to help the children studying in government schools in a much better way towards their post secondary education.
4. The model proposed as an academician can be useful to build a software model to provide a solution by formulating the result.
Cite This Work
To export a reference to this article please select a referencing style below:
