Warehousing and Analysing Educational Data
- Speaker: Dr Doulkifli Boukraâ, University of Jijel
- Date: Thursday, 4 February 2016 from 13:00 to 14:00
- Location: Room 151
The goal of our work is to provide the decision makers (lecturers, heads of teaching programs, deans, etc.) at the University of Jijel (Algeria) with a decision support system that helps them make better decisions in the educational field. To that purpose, we proposed a warehousing system composed of a central data warehouse, a set of data sources and a set of front-end applications to restore and analyse the data. In this talk, I will present the data warehouse schema that relies on a multidimensional model presented at different levels. The multidimensional model uses particularly parameters to evaluate and assess the students’ performance as well as the lecturers’ efficiency. The front-end applications are based on the following data mining techniques.
- Discovering association rules between the performance of students on the one hand and the courses along the years of their studies and across different cycles (bachelor, master, doctorate) on the other hand.
- Predicting the students’ performance by using decision trees.
- Discovering the students’ profiles by using clustering techniques and analysing the accordance of the profiles with the corresponding specialty of study.
Biography: I obtained my Engineer degree (1998), my Magister degree (2005) and my PhD (2013) from the Ecole Superieure d’informatique (www.esi.dz). Currently, I am an associate professor and researcher at the University of Jijel where I am responsible of the ISDS (information systems and decision support) master’s programme. My research interests lay in multidimensional modelling of (complex) data warehouses, XML warehouses, performance optimization of data warehouses, educational data warehousing and educational data mining.