While mashup tools mainly focus on modeling the data flow and rapid development, model-driven engineering approaches permit different views and more expressive modeling concepts.
Greater complexity means more expressive models.
Our challenge is to design an expressive model of gesture aiming at characterizing each frame of a gestural sequence.
The experimental results show that in many cases ensembles of logistic regression classifiers may outperform more expressive models due to their robustness to noise and low sample density in a high-dimensional feature space.
So far the model appears still to be manageable in other less expressive modeling approaches.
We are looking to improve on this by using a slightly more expressive model, such as, non-linear SVM model.
We believe a more expressive model that is able to model such interactions between coverage terms would improve on the current results.
They show that ontologies fulfill most of these requirements and are the most expressive models as these formal models allow the integration and exploitation of more specific context knowledge with high-level context information using reasoner components.
To provide protein engineers and scientists with comprehensive information and a more expressive model, we populate our domain ontology with the extracted information, which can then be queried as a knowledge base.
