Deliverable D5.6

[cbadenes]

Techniques of personalized recommendations of ROs with reports

T5.4 Personalised recommendations of ROs

We will focus on the adaptation of existing collaborative filtering algorithms to the Dr Inventor scientific creativity context. We will (a) consider complex models in Collaborative Filtering algorithms, so that they do not only work with single ratings or values to characterize an object (e.g., a 1 to 5 rating for a workflow or data source), but with more structured models (from a set of values for the different attributes of the Research Object to a more complex knowledge model about the contents of the Research Object from the scientist’s point of view) and with information regarding their evolution, and(b) include ontology matching techniques (T5.3) to bridge across different knowledge models, so that initial differences between knowledge models are smoothed and similarities can be found. (c) integrate ontology reasoning in traditional Collaborative Filtering algorithms, so that additional similarities and differences can be obtained from those currently obtained with traditional techniques.

T5.5 Evaluation methods and tools

An evaluation methodology will be put in place for the semantics tasks. We will integrate whenever possible available tools or implement our own software. The task will also run the different intrinsic evaluations which will help assess the progress of the technological development in the work package.