DataBearings

Previous year's Nominees

DataBearings

https://sites.google.com/site/databearings/

DataBearings is a very fast and lightweight semantic rule and semantic programming engine that also allows virtualizing external federated non-semantic data.

Essentially, DataBearings allows lifting distributed non-semantic data as Linked Data for further integration, processing, and reasoning. The source data can be in relational databases, Web services, files, etc. and is accessed on-demand and does not need to ever be replicated and stored as RDF.

The DataBearings approach is very innovative. DataBearings is the only semantic data virtualization solution available at present,
which is not emulating virtualization via query-time Extract-Transform-Load to RDF, while capable of working with Web data and not only relational databases.
It is also the only semantic solution to the data integration problem that is as fast as non-semantic ones and so lightweight that can in principle be run on an Android smartphone.
It also provides some unique features, it allows: mapping data to arbitrary semantic structures as dictated by target ontologies, performing source data transformations prior to mapping to an ontology, parametrization of data source requests, and flexibly combining virtualized non-semantic and locally stored semantic data.

DataBearings has been applied in a smart home pilot system as described in
https://www.linkedin.com/pulse/my-smart-home-head-cloud-artem-katasonov
and in a couple of business systems in the parking domain, used by the customer, Finnpark, in their day-to-day operations now, as described in
https://www.linkedin.com/pulse/tampere-city-most-integrated-parking-data...
The latter case demostrates the technological maturity of DataBearings, while both cases together demonstrate its potential in multiple different domains.

In addition to the DataBearings execution platform, DataBearings IDE for Eclipse has been under development.
DataBearings IDE focuses on automating the development of data virtualization and integration cases, including via the application of ontology mapping tools,
making the implementation of simple cases extremely fast and comprehensible for the users, as can been in the tutorial video
https://youtu.be/PV2MI7raLjQ

DataBearings is a part of the Finnish national Tekes-funded Internet of Things program http://www.internetofthings.fi/
and builds on results from an earlier Finnish Tekes-funded project.

VTT

Artem Katasonov, Senior Scientist at VTT Technical Research Center of Finland

Tekniikankatu 1
33720 Tampere , LS
Finland