The LIMES development team is happy to announce LIMES 1.5.0!
LIMES is a link discovery framework for the Web of Data, which implements time-efficient approaches to discover links among POI resources. LIMES attempts to harness similarity measures in many dimensions including (but not limited to) the euclidean, spatial and temporal dimensions. Our framework provides a cross-platform graphical user interface as well as several machine learning algorithms which enable an easy configuration of the link discovery tasks.
In the course of the SLIPO project, we mainly focussed on improving the efficiency and performance of link discovery based on spatial and temporal relationships. To this end, we developed the novel approach of RADON for rapid temporal relation discovery as well as the AEGLE algorithm for the optimized discovery of temporal relation. Moreover, in order to assure the quality of the generated links among POI resources, we implemented a new functionality in LIMES 1.5.0 to generate a number of quality indicators such as runtime and F-measure. Such quality indicators help the LIMES user to quickly evaluate the result of his/her link discovery task.
This is what happened since our last update regarding LIMES:
- Reworked LIMES Server REST API
- Rewrite of the preprocessing package which allows for novel complex preprocessing functions
- Improvement to evaluation package
- Implemented quality indicator feature
- Restructured user manual to allow for a quick start with LIMES
- Updated Dockerfile
- Several bug fixes