Christian Grevisse

An Extensible and Lightweight Modular Ontology for Programming Education

Semantic web technologies such as ontologies can foster the reusability of learning material by introducing common sets of concepts for annotation purposes. However, suggesting learning material from an open, heterogeneous corpus is a nontrivial problem. In this paper, we propose an extensible and lightweight modular ontology for programming education. Its main purpose is to integrate annotated learning material related to programming into an IDE such as Eclipse. Our ontology is based on a modular architecture, which is extensible with respect to different programming languages. Aligning language-specific concepts with user-specific tags allows us to suggest learning resources for code elements in a fine-grained and cross-curricular way. Our concrete implementation establishes relations between learning aspects in Java or C code and annotated resources such as articles on online question-and-answer sites.

Knowledge Graph-Based Teacher Support for Learning Material Authoring

Preparing high-quality learning material is a time-intensive, yet crucial task for teachers of all educational levels. In this paper, we present SoLeMiO, a tool to recommend and integrate learning material in popular authoring software. As teachers create their learning material, SoLeMiO identifies the concepts they want to address. In order to identify relevant concepts in a reliable, automatic and unambiguous way, we employ state of the art concept recognition and entity linking tools. From the recognized concepts, we build a semantic representation by exploiting additional information from Open Knowledge Graphs through expansion and filtering strategies. These concepts and the semantic representation of the learning material support the authoring process in two ways. First, teachers will be recommended related, heterogeneous resources from an open corpus, including digital libraries, domain-specific knowledge bases, and MOOC platforms. Second, concepts are proposed for semi-automatic tagging of the newly authored learning resource, fostering its reuse in different e-learning contexts. Our approach currently supports resources in English, French, and Spanish. An evaluation of concept identification in lecture video transcripts and a user study based on the quality of tag and resource recommendations yielded promising results concerning the feasibility of our technique.

Visualization and classification of protein secondary structures using Self-Organizing Maps

In molecular biology, it is estimated that there is a correlation between the secondary structure of a protein and its functionality. While secondary structure prediction is ultimately possible in wet lab, determining a correlation with the functionality is a hard task which can be facilitated by a computational model. In that context, this paper presents an automated algorithm for the visualization and classification of enzymatic proteins with the aim of examining whether the functionality is correlated to the secondary structure. To that end, up-to-date protein data was acquired from publicly accessible databases in order to construct their secondary structures. The resulting data were injected into a tailored version of a Kohonen Self-Organizing Map (SOM). Part of the work was to determine a proper way of reducing large secondary structures to a common length in order to be able to cope with the constant dimensionality requirement of SOMs. The final contribution consisted in the labeling of the trained nodes. Eventually, we were able to get a visual intuition and some quantified assessment on the nature of this correlation.