Semantics driven data analytics is a subdomain of data analytics and makes use of semantic sources and semantic Web technologies to enable decisions based on data that is human and machine readable alike. The transformation of numerical data is performed in a way that a human can interpret the information and – the other way around – humans produce those data resources, e.g. ontologies, that help to interpret the numerical data. Altogether, the existing information becomes embedded into a representation that is well aligned with our human understanding of our observations, and the experimental observations can be aligned with our human understanding exploiting computer-based approaches.
In biomedical research, different disciplines in science work together to advance science at an unprecedented pace based on existing big data approaches. This leads to the result that clinicians exploit results from basic biomedical research, e.g. the functioning of genes, and medicinal chemists work together with chemists and molecular biologists to produce new drugs. The existing collaborative work hinges around large data resources that are aligned according to their relevance for diseases, functions and procedures in one or many organisms, and the automatic analysis of large data resources based on our knowledge representation. Without semantic technologies, computers would less capable to support decisions in a human readable way.
Data analysis is a research domain showing a high diversity in data usage. The shared goal is the prediction of outcomes based on data and knowledge discovery as a means to identify findings that are well-supported by the data and showing specific relevance that has not been recognized yet.
Methods in biomedical data analytics and semantic analytics comprise statistics, machine learning on experimental data, preparation of semantic resources and large-scale data integration using semantic Web technologies.
This resource has been funded by Science Foundation Ireland under Grant No. SFI/08/CE/I1380 (Lion-2) and by Grant No. SFI/12/RC/2289 INSIGHT – National University of Ireland, Galway