Wed, Oct. 28 4:00 PM-5:00 PM
Lauri Lahti, Helsinki University of Technology, Finland
We propose a new method for guided generation of concept maps from open access online knowledge resources such as Wikies. Based on this method we have implemented a prototype extracting semantic relations from sentences surrounding hyperlinks in the Wikipedia’s articles and letting a learner to create customized learning objects in real-time based on collaborative recommendations considering her earlier knowledge. Open source modules enable pedagogically motivated exploration in Wiki spaces, corresponding to an intelligent tutoring system. The method extracted compact noun–verb–noun phrases, suggested for labeling arcs between nodes that were labeled with article titles. On average, 80 percent of these phrases were useful while their length was only 20 percent of the length of the original sentences. Experiments indicate that even simple analysis algorithms can well support user-initiated information retrieval and building intuitive learning objects that follow the learner’s needs.