期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2015
卷号：4
期号：10
页码：9823
DOI：10.15680/IJIRSET.2015.0410072
出版社：S&S Publications
摘要：Educational data mining is an emerging trend, concerned with developing techniques for exploring, andanalyzing the huge data that come from the educational context. EDM is poised to leverage an enormous amount ofresearch from data mining community and apply that research to educational problems in learning, cognition andassessment. In recent years, Educational data mining has proven to be more successful at many of these educationalstatistics problems, due to enormous computing power and data mining algorithms. The main objective of highereducation institutions is to provide quality education to its students. One way to achieve highest level of quality inhigher education system is by discovering knowledge for prediction about students‟ performance. This study presentsan approach to predict the performance of students in a semester exams. This approach is based on a Naive Bayesclassifier. The objective is to know what grades students may obtain in their end semesters results. This helps theeducational institute, teachers and students i.e., all the stakeholders involved in an education system. They can bebenefitted from the prediction of students‟ results in multitude of ways. Students and teachers can take necessaryactions to improve the results of those students whose result prediction is not satisfactory. A training dataset of studentsis taken to build the Naive Bayes model. The model is then applied on the test data to predict the end semester resultsof students. In this study, number of attributes is considered to predict the grade of a student.
关键词：Student performance; Naive Bayesian classifier; Educational data mining; Prediction.
