Authors : S. Mythili, R. PradeepKumar and P. Nagabhushan
Abstract: Recent developments in database technology have seen variety of data being stored in huge collections which are being referred to as generic database. The wide variety of data makes the analysis tasks of a generic database a strenuous task in Knowledge Discovery (KD). One approach to simplify the Strenuous Task of D is to summarize large data sets in such a waythat the resulting summary dataset is of manageable size and yet retains as much of the knowledge in the original dataset as possible. This process is termed as Symbolic Data Analysis (SDA). In SDA, Histogram has received significant attention as summarization object. This study demonstrates the approach of Symbolic Data Analysis (SDA) in very large database . This methodology analyzes large, very large datasets and extract glean useful information from within their massive confines. In this study, SDA uses histogram to summarize the data and linear regression to approximate the histogram. The application of SDA is illustrated in education environment using LMS Quiz data obtained from Ekluv-Ya . It discovers knowledge pattern in the dataset along with the help of data mining technique clustering. The application of this framework can serve as an assistance tool for managers of higher education institutions in improving the educational quality level.
S. Mythili, R. PradeepKumar and P. Nagabhushan, 2016. Knowledge Discovery in Big Data Using Symbolic Data Analysis. Asian Journal of Information Technology, 15: 3943-3948.