RT1: Semantic annotation of texts
Use Hybrid AI to semantically annotate large text collections that cover vast historical time periods, revealing patterns and biases. The objective is to produce open data that are shareable, searchable, findable, and linkable to external resources.
RT2: Classifying and naming objects
Describe (knowledge modelling), classify (using machine learning to analyze images) and name objects (terminology and standards) in order to produce shareable and searchable datasets in Linked & Open Data formats for the Semantic Web.
RT3: Pattern recognition in literary/journalistic/other corpora
Using NLP, Knowledge Representation (KR) and Deep Neural Networks for pattern detection in order to detect linguistic and semantic patterns in a large corpus.
RT4: Digitalization and datafication of education
Training AI armed by the Knowledge Space Theory, will help us to define and represent students’ knowledge gaps, taking into account the complexity of scientific concepts’ relations between each other.