Flexible View Definitions to Enhance E-­‐learning Resources Availability

Jalel Akaichi

Abstract


An E-learning Data Warehouse (EDW) is constituted of information collected from heterogeneous, distributed, and autonomous E-learning Information Sources (EISs). EDW, which is fed by view definitions built upon EISs, is frequently queried by e-learning actors such as educators, learners and decision makers for many reasons going from analysis to mining. The above view definitions, which represent considerable informational resources, can become undefined when EISs change their schemas accordingly to their autonomy. This obviously decreases e-learning resources availability and consequently affects analysis and mining efficiency. In this paper, we propose to study the issues of using agents based architecture to achieve e-learning data warehouse maintenance under schema changes by repairing automatically affected view definitions. This implicitly optimizes e-learning resources availability by automatically finding replacements for affected components belonging to view definitions and representing critical information.



Back to list of accepted papers