FAN Xiaoliang

PhD graduated
Departure date : 07/19/2012

Supervision : Patrick BRÉZILLON

Context-oriented Scientific Workflow and its Application

The lack of flexibility is an inhibitor in the use of scientific workflow (SWF) systems to transform hypotheses into scientific discoveries.However, notable absences of methodology for modeling SWF in a uniform representation are seen , which results in the lack of user-oriented learning processes. A real use of SWF requires the introduction of the concept of “context”. Current failures are attributed to the lack of explicitness of context in SWF models, the insufficiency in making adaptive decisions, the incomplete SWF sharing mechanism. To address these problems, we proposed a Context-oriented Framework for Scientific WorkFlow (CoF4SWF) to make contexts explicit by Contextual Graphs in SWF lifecycle. We implemented CoF4SWF methodology by a functional prototype called Context-oriented SWF (CxSWF). Moreover, CxSWF prototype was testified through a real world application “virtual screening research on avian influenza H5N1 virus”. Finally, the results from twofold evaluation processes illustrated that initial hypotheses of the virtual screening application have been reached in principle. The main contributions of the thesis are: (i) A context-oriented methodology for studying SWF modeling; (ii) The identification of three approaches of making contexts explicit in SWF models; (iii) The investigation on the role of contextual graphs in SWF modeling. The originality of the thesis stands on: (1) Representing and modeling SWF flexibility with its contexts to enable adaptive decision making; (2) Making contexts as the status of knowledge to be built up for practice learning; and 3) Sharing of SWF model with its contexts as a complementary output for scientific discoveries.

Defence : 11/07/2011 - 14h - Chine

2010-2017 Publications