DOP8_Cycle: Merging both data and analysis operators life cycles for Technology Enhanced Learning
Résumé
This paper presents the DOP8_Cycle, a Data Mining Iterative Cycle that improves the classical data life cycle. While the latter only combines the data production and data analysis phases, the DOP8_Cycle also integrates the analysis operators life cycle (a combination of both the analysis operator design and data analysis phases). Considering these two life cycles in the same environment leads to fostering the link between data and operators and to creating analysis processes for a wide range of Technology Enhanced Learning (TEL) research. This concept is reified in a new computing platform, called UnderTracks. The latter provides a high flexibility on storing and sharing data, operators and analysis processes. Several real TEL analysis scenarios are now present into the platform, and we analyze them (1) to test the DOP8_Cycle efficiency, (2) to test Undertracks flexibility on storing data and operators and (3) to test Undertracks flexibility on designing analysis processes. This evaluation shows that the DOP8_Cycle allows data, operators and process analysis to be shared between different TEL domains.