In our work we describe an approach for the automatic evaluation of student test results by a web based e-learning application, which considers more criteria than the obvious correctness. This evaluation is focused on the use in adaptive learning paths, meaning that the student's learning path is adapted based on the test results. We describe a concept which integrates the evaluation of more criteria than correctness and how an implementation of such adaptive learning paths could look like in a web based e-learning environment. Our work shows a working prototype which uses the enhanced criteria speed and certainty based upon an evaluation with fuzzy logic algorithms. Finally we present some first results about the acceptance of web based adaptive learning paths.

In our work we describe an approach for automated and approximate evaluation of character based answers in computer-based tests. Cloze exercises (`fill-in-the-blank') are error-prone to a variety of potential error sources and are subject to syntactic and semantic ambiguities. Our approach combines string similarity measurements with semantic word analysis to render this type of computer-based exercise more robust and user-friendly. We present a working prototype for a web-based e-learning system supporting spelling correction (syntax level) and an extension to synonyms (semantic level). A community-based approach using OpenThesuarus.org is proposed to gather and evolve missing synonyms for given answers of cloze exercises.

The thirst for information in complex working environments calls for intelligent systems which optimally assist the user, i.e. which offer the user the most relevant information. The aim is to decrease the time the user has to spend on his hunt for information and to offer him the best fitting help and learning material in an on-demand manner. We present an approach for the semantic retrieval of help and learning material which takes the working context into account. Based on the semantic structure of an ontology with attached binding weights a context-aware ranking of help and learning material is generated. The semantic search results fit better to the learner's actual situation than e.g. a pure full-text search, because the underlying ontology-based retrieval is aware of relations in the search domain and uses this knowledge in a way aligned to the learning process as well as to the specific domain. The results of the semantic search are presented for an application scenario in radar-based image interpretation. The advantages of the semantic approach are shown by a comparison with a state-of-the-art full-text search engine.