A Decision Support System for Course Offering in Online Higher Education Institutes

Ahmad A. Kardan, Hamid Sadeghi
2013 International Journal of Computational Intelligence Systems  
Prior to every academic semester, every department's administrator is required to offer the best overall set of courses to meet student requirements, instructor needs and department regulations. The key contributions of this research is firstly, determining the potential factors that influence student behavior on the online courses they choose, secondly, modeling the course offering problem and fitting a function to a training set of data using neural network approach, thirdly, design and
more » ... entation of a decision support system to help the department's administrator to simulate student behavior in course selection process and support his/her decisions on the courses to be offered, and lastly, employing the proposed decision support system to perform what-if analysis and goal seeking behavior. The samples of the experiments came from 298 online graduate courses in 14 academic terms from 2005 to 2011. The results revealed high prediction accuracy on the experimental data. The performance of the introduced decision support system was also compared with three well-known regression techniques, "support vector regression", "K-nearest neighborhood", "decision tree" and a traditional approach. The finding exposed that the suggested decision support system outperformed the others significantly. the complex nature of these kinds of problems, it is a difficult task for an individual to select the best alternative from the existing solutions. Modern computerized systems have capabilities that can facilitate human decision making in a number of ways, including; speedy computations, improved communication and collaboration, increased productivity of group members, improved data management, overcoming cognitive limits in processing and storing information, quality support, agility support, using the Web, and also anywhere and anytime support 47 .
doi:10.1080/18756891.2013.808428 fatcat:lzpc4ub2afhntmqhy5w3usonma