LIMES – LInk discovery framework for MEtric Spaces
From the website:
LIMES is a link discovery framework for the Web of Data. It implements time-efficient approaches for large-scale link discovery based on the characteristics of metric spaces. It is easily configurable via a web interface. It can also be downloaded as standalone tool for carrying out link discovery locally.
LIMES detects “duplicates” in a single source or between sources by use of string metrics.
The current version of LIMES supports exclusively the string metrics Levenshtein, QGrams, BlockDistance and Euclidian as implements by the SimMetrics library. Further metrics will be included in following versions.
An interesting approach to use as a topic map authoring aid.
Questions:
- Using the online LIMES interface, develop and run five (5) link discovery requests. Name and save the result files. Upload them to your class project directory. Be prepared to discuss your requests and results in class.
- Sign up to be discussion leader for one of the algorithms supported by LIMES. Prepare a two (2) page summary for the class on your algorithm.
- What suggestions would you have for the project on its current UI?
- Use LIMES to augment your topic map authoring. Comments? (3-5 pages, no citations)
- In an actual run, I got the following as owl:sameAs – http://bio2rdf.org/mesh:D016889 and http://data.linkedct.org/page/condition/4398. Your evaluation? You may follow any links you find to make your evaluation. (2-3 pages, include URLs for other locations that you visit)
[…] Levenshtein distance test mentioned in the LIMES post is an example of a string matching […]
Pingback by SecondString « Another Word For It — November 15, 2010 @ 4:56 am