SAXEF: A System for Automatic eXtraction of learning object Features

New online courses are often created by using existing materials on the net.However, those materials are usually proposed without information on their aimsand the typology of users which they are destined to. Moreover, the contents arenot clearly synthesized so that a complete analysis of the whole materials is oftennecessary to understand their relevance to the new course.Using our experience on the creation of online courses with existing web materials,we have thought how to help teachers in finding the best materials for the creationof new online courses. To this end, we have developed a system, called SAXEF(System for Automatic eXtraction of lEearning object Features), that is capable toautomatically extract the didactic indicators (a sort of DNA) of any web page (orgroup of pages) found on internet and allows a teacher to easily evaluate whetherthat page (with its contents) is of interest to him/her. This paper presents the mainarchitecture of SAXEF, its implementation details and some experiences on its use.At present, SAXEF is capable to automatically extract various didactic indicatorssuch as main and secondary topics, synthesis level and multimedia level from anygroup of web pages.