2016-2020

Description:
This project will improve the state of the art in semantic data mining by establishing novel methods and algorithms for the automated semantic annotation and enrichment of data, whose output will be explored by novel data mining approaches capable of capitalizing on the semantic web.
In particular we will make contributions in a number of areas:
1. Semantic Data Annotation: where we will develop methods for the automated selection of ontologies and the annotation of data elements using multiple ontologies;
2. Ontology Matching: where we will build upon or ontology matching system to tackle the challenge of building hyperontologies combining multiple ontologies from distinct but related domains;
3. Semantic Data Enrichment: where we will develop strategies to enrich the data with information extracted from the Linked OpenCloud;
4. Semantic Similarity: where we will develop novel semantic similarity measures tailored to serve as a distance metric over the multi-domain hyperontology;
5. Semantic Data Mining: where we will design novel data mining approaches that are able to explore the semantic data network.
This is an emerging field of research, with as yet no established methodologies, and only a few tentative approaches published that are unable to combine the expressiveness of the semantics encoded in ontologies with the multitude of data available as LOD.

Funding Entity:
FCT

Reference:
PTDC/EEI-ESS/4633/2014