Evaluation of Student Performance: An Outlier Detection Perspective
P. Ajith1, M.S.S. Sai2, B. Tejaswi3
1P. Ajith, Research Scholar, KL University, India.
2M.S.S. Sai, Professor. Department of CSE, KKR & KSR Institute of Technology & Sciences, Guntur, A P, India.
3B. Tejaswi, Asst. Prof. Department of CSE, KKR & KSR Institute of Technology & Sciences, Guntur, AP, India
Manuscript received on 09 January 2013 | Revised Manuscript received on 18 January 2013 | Manuscript Published on 30 January 2013 | PP: 40-44 | Volume-2 Issue-2, January 2013 | Retrieval Number: B0361012213 /2013©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Educational data mining is current growing research area and the main essence of data mining concepts are used in the educational field for extracting useful information of the students based on their behavior in the learning process. Prior approaches used decision tree classifications optimized with ID3 algorithms to obtain such patterns but discovering the implicative tendencies is valuable information for the decision-maker which is absent in tree based classifications. So we propose to use outlier detection for mining and evaluating educational data of students. In this paper, outlier detection mechanisms are used for identifying outliers which improve the quality of decision making. We used outlier analysis to detect outliers in the student data. In proposed system, clustering mechanism along with univariant analysis is implemented. Clustering is finding groups of objects such that the objects in one group will be similar to one another and different from the objects in another group. While clustering, the large data set is divide into clusters which consists of outliers. After Clustering, the data points which are present outside the clusters are identified and treated as outliers. Identification is done by using univariate analysis which is the simplest form of quantitative (statistical) analysis. A basic way of presenting univariate data is to create a frequency distribution of the individual cases Here, we analyze the performance of UG students of our college and present the results using outlier detection mechanism. The analyzed results are represented using histograms which are based on univariate analysis.
Keywords: Outlier, Clustering, Univariate analysis, and Histograms
Scope of the Article: Smart Learning and Innovative Education Systems