[
https://issues.apache.org/jira/browse/COMMONSRDF-61?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16259927#comment-16259927
]
Aaron Coburn commented on COMMONSRDF-61:
----------------------------------------
This issue seems related to commons-text and not commons-rdf. Perhaps this
issue can be closed here and reopened in the TEXT project if it is still
relevant.
> Incorrect test data provided in JaroWinklerDistance algorithm project
> ---------------------------------------------------------------------
>
> Key: COMMONSRDF-61
> URL: https://issues.apache.org/jira/browse/COMMONSRDF-61
> Project: Apache Commons RDF
> Issue Type: Improvement
> Components: api
> Reporter: Christopher Stoneley
> Priority: Minor
> Labels: easyfix, patch, test
> Original Estimate: 5m
> Remaining Estimate: 5m
>
> Imported from org.apache.commons.text.similarity.SimilarityScore;
> Test data provided from the following URL has been proven to be incorrect:
> https://commons.apache.org/sandbox/commons-text/jacoco/org.apache.commons.text.similarity/JaroWinklerDistance.java.html
> Line 69 displays the returned value 0.95 from calling the function:
> distance.apply("D N H Enterprises Inc", "D & H Enterprises, Inc.") = 0.95.
> After testing this data extensively across multiple languages the resulting
> data from this function should result in 0.85.
> The rest of the test data has been proven to be correct.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
