Abstract: Formal Concept Analysis (FCA) developed in the last 30 years to a prominent field of Knowledge Discovery, Processing and Representation. Well-known for its intuitive representation of conceptual structures as order diagrams, FCA handles conceptual hierarchies which can be derived from various data sets. This makes FCA suitable for knowledge management tasks in many fields. We consider the problem of using FCA as a quality of education evaluation tool for the instructor. Students are using educational resources in an e-learning platform and the raw data set consists only of the subsequent recorded weblogs. After a preprocessing stage, various conceptual structures – among them behavioral patterns of e-learning resources usage – are distilled from the dataset using conceptual scaling and triadic FCA and are altogether offered as a decision support for the instructor. We prove the efficiency of using FCA in comparing students behavior through the educational process and highlight how new educational strategies correlate with their academic performance and the use of online learning ressources.
Keywords: Formal Concept Analysis, Knowledge Discovery, e-learning platform, triadic FCA, educational process
Acknowledgments: Diana Halita was supported by a doctoral research from POSDRU/187/1.5/S/155383.
Presentation: SOFTCOM 2016 – Session 1: Education and Information Systems
