EFFICACY AND EFFICIENCY OF EDUCATIONAL DATA MINING THROUGH MULTISTRATEGY MACHINE LEARNING

Abstract

The present day higher educational institutions in the country face the tough challenge of surviving in the academia and producing cutting edge professionals for an industry with ever growing demand of meeting the dynamically changing requirements. The absence of a comprehensive data analysis framework and combined analysis of correlated factors affecting the teaching learning process may give rise to false interpretations which are relied upon by the institutions. As a resultant, the system may fail to take effective measures or introduce necessary interventions in time to handle the deviations in the right manner thus defeating the purpose of meeting the program outcomes. In the absence of an accurate comprehensive quality management mechanism individual perceptions bring subjectivity into the quality evaluation and so the results are skewed and give rise to false projections or fabricated interpretations. Major decisions of regulating bodies or institutes themselves are dependent on such analyses which necessitate correct conclusions. Thus quality management of teaching learning process is of utmost importance from the perspective of student, teacher and all other stake holders. It provides a scientific unambiguous basis for quality evaluation and consequently better quality management. A multi strategy machine learning framework goes a long way in supporting evaluation of both qualitative and quantitative aspects by reducing uncertainty in data analysis and facilitating effective quality management.