Work place: BVB college of Engineering and Technology, Hubli, Karnataka, India
Research Interests: Computational Science and Engineering, Computational Engineering, Engineering
Nitin Kulkarni has a B.S. degree in Mechanical Engineering from Karnataka University Dharwad, India in 1984. He holds a MBA in Human Resource Management from Visvesvaraya Technological University, Belgaum India in 2010. From 1984-2002, he worked in industries ranging from Machine tools, Aerospace, Tool and Die, Software, Consumer Electronics. His last Industry job was at Microsoft Corporation, Redmond, WA, USA, as a Group Engineering Manager responsible for New Hardware Product development. His academic career started as a lecturer in 2002 during which he took the responsibility of placements at SDM College of engineering and Technology, Dharwad, India. Currently he is the Director at Center for Technology Innovation and Entrepreneurship at BVB college of Engineering and Tech, Hubli, India, and is also an Associate Professor at the School of Management Studies and Research (SMSR) at BVB Hubli. His research interests include Measuring and enhancing Employability of Fresh Engineering Graduates of North Karnataka region, Entrepreneurship and its impact on enhancing employability.
DOI: https://doi.org/10.5815/ijmecs.2016.02.08, Pub. Date: 8 Feb. 2016
Data Mining is a dominant tool for academic and educational field. Mining data in education atmosphere is called Educational Data Mining. Educational Data Mining is concerned with developing new methods to discover knowledge from educational/academic database and can be used for decision making in educational/academic systems. This work demonstrates an effective mining of students performance data in accordance with placement/recruitment process. The mining result predicts weather a student will be recruited or not based on academic and other performance during the entire course. To mine the students’ performance data, the data mining classification techniques such as – Decision tree- Random Tree and J48 classification models were built with 10 cross validation fold using WEKA. The constructed classification models are tested for predicting class label for new instances. The performance of the classification models used are tested and compared. Also the misclassification rates for the classification experiment are analyzed.[...] Read more.
Subscribe to receive issue release notifications and newsletters from MECS Press journals