csarven / transparency-linked-data Goto Github PK

Make sure the RDF documents that are currently in the CPI2012 branch are valid.

If you create new URI patterns, make sure to provide documentation for it.

Can I create a new branch for CPI2013 or is this up to the owner of the repository? I just don't want to mess things up or break something ;)

The CPI2012 is calculated by subtracting the mean of the data (13 sources) and dividing them by the standard deviation and then rescaled to have a mean 45 and standard deviation 20..

I searched for a term for this calculation on dbpedia, but could not find anything. Do you have a hint where I can possibly find a term for such a calculation?

If you run the file tarql.mappingCPI2012.sparql from the branch CPI2012 with tarql you will see that it will create all triples, including the ones where the score from the surveys are equal to 0.
So the output is as follows (example for African Development Bank Governance Ratings, Limited to 1 set of triples):
...
http://transparency.270a.info/dataset/corruption-perceptions-index/2012/sources/african-development-bank-country-policy-and-institutional-assessment-governance-ratings/2012/DK
rdf:type qb:Observation ;
rdfs:label "Corruption Perception Index 2012 for Denmark from African Development Bank Governance Ratings in 2012" ;
dcterms:source http://cpia.afdb.org/ ;
qb:dataSet "http://transparency.270a.info/dataset/corruption-perceptions-index/2012/sources" ;
sdmx-attribute:adjustment
http://transparency.270a.info/classification/attribute/matching-percentiles ;
sdmx-dimension:refArea
http://transparency.270a.info/classification/country/DK ;
sdmx-dimension:refPeriod
http://reference.data.gov.uk/id/year/2012 ;
property:score "0"^^xsd:decimal ;
property:source http://transparency.270a.info/classification/source/african-development-bank-country-policy-and-institutional-assessment-governance-ratings/2012 .
...
You see that these triples should not be generated, since the property:score is equal to "0"^^xsd:decimal ; --> e.g There was no survey for Denmark from African Development Bank.

I tried to exclude these triples with the following:

• MINUS{?observationAFDB property:score "0"^^xsd:decimal} //this didn't work
• FILTER(?scoreAFDB < 1) //this filtered everything e.g --> this filtered every triple, which was generated for Denmark (also triples where there were actually surveys)
• and OPTIONAL{} //this had no impact

Do you have a hint or idea how I can approach this? Thx

For instance, if the logs from your local machine are not required, do not commit them.

You told me to crate the dsd file for the CPI2102. So I looked at the meta.ttl file in the data folder of the master branch. So is it the aim to extend this meta.ttl file for the CPI2012 or create a new one?

Second question:
I see, that it says which URI represent what (with comments and properties). But I can't see the final propose of the file. Can you explain me what this file finally serves for?