Robin Nicolay

Autonomous Semantic Structuring of Lecture Topics - Synthesis of Knowledge Models.

Students attending lectures in universities suffer from a weak structural awareness on lecture content. According to learning theories, structural awareness is a relevant factor to association and comprehension of new learning inputs. We synthesize semantic structures from non annotated lecture slides using Topic Modeling algorithms to identify relevant terms and relate them in force-directed graphs. The synthesized graphs provide a structural overview on the topic distribution and relations of non annotated sequential lecture slides.

SYNTHESIS OF PEDAGOGICAL ANNOTATIONS

For the creation of learning session in an ITS pedagogical knowledge is needed. This is created by teachers. An ITS (Intelligent Tutoring System) should have some abilities of teachers, e.g. for learning content creation. It is the future of ITSs to be more independent from human pre-defined content. This would allow an ITS to reduce the workload of teachers in adapting content to a student through automatic creation of learning content. This paper introduces a concept of different layers of information to be used with a rule system with which annotations in an educational context for expert knowledge can be created. The level of automation is one step beyond the classical ITS concept.

Evaluation of acoustical position determination in a classroom scenario

This study evaluates whether ultrasound tones can be used to determine position of mobile devices in classrooms. The aim is to use this position to find the seating scheme of students for another study in the area of e-learning. We have conducted several experiments to test mobile devices for their suitability. We have also performed experiments in real scenarios and implemented a software system that for position determination. We measure distances and apply the least-square algorithm to calculate the position of mobile devices.

Agent-Based Proactive Support in Smart Environments

We describe an agent based approach to control a heterogeneous and dynamic device ensemble. The resulting controller provides pro-active assistance to the users of the environment. For this, roles defining the functionality are attached to dynamically constructed agents. Those roles are assigned based on the current situation, that is the accessible state of all devices and the currently assigned roles. The feasibility of the approach is proven by applying it to a non-trivial use case of a smart meeting room. Finally we present the results of a user evaluation.