- From: Debattista, Jeremy <Jeremy.Debattista@iais-extern.fraunhofer.de>
- Date: Mon, 14 Jul 2014 19:07:29 +0000
- To: Antoine Isaac <aisaac@few.vu.nl>
- CC: Public DWBP WG <public-dwbp-wg@w3.org>
Hi Antoine, > I have a question though: is there any existing daQ instantiation to measure 'general quality' of datasets, that we could have a look at? > Something like what is presented in Figure 2 of your paper, but a more systematic… We have a number of daQ instantiation to measure quality [1] - most of them are not generic to all datasets though. Cheers, Jer [1] https://raw.githubusercontent.com/diachron/quality/master/src/main/resources/vocabularies/dqm/dqm.trig On 11 Jul 2014, at 17:39, Antoine Isaac <aisaac@few.vu.nl> wrote: > Hi Jeremy, > > Thanks a lot for the contribution! > > Right now we're still at discussion stage, I'm afraid. > As we discussed yesterday, I like very much your idea of using a generic metrics framework (like daQ) to express the quality dimensions that are important for us, as per our use cases and requirements. > Of course this assumes that we will identify the dimensions, and appropriate metrics to evaluate datasets against them. Right now we're not there. > > I have a question though: is there any existing daQ instantiation to measure 'general quality' of datasets, that we could have a look at? > Something like what is presented in Figure 2 of your paper, but a more systematic... > > Best regards, > > Antoine > > On 7/10/14 6:58 PM, Debattista, Jeremy wrote: >>> Hi Antoine, >>> >>> I hope this email finds you well. I would like to inform you that we have extended our daQ ontology with the Data Cube vocabulary so that we could now represent “observations” on quality. We have a paper about this accepted at the main conference at semantics[1]. I’ve attached the paper for you, but probably I will do some minor changes for the camera-ready version. >>> >>> >From what I’ve understood till now, the efforts the WG is doing is different from the scope of daQ. The daQ is a generic framework allowing (1) the uniform representation of metrics; (2) suggests how quality metadata is represented. On the other hand, the WG would like to identify those domain-independent quality metrics which each Linked Dataset can be measured to. Thus, the definition of these metrics will be defined in an ontology (such as DCAT). >>> During the last call, you mentioned something about the work I am doing. The daQ can support the definition of these quality metrics (say in the DCAT vocab) by having the definition of these metrics extending daQ. We are already defining a number of specific metrics in an ontology for the EU project I am working on [2] (this is also explained in the attached paper). >>> >>> Also, I am wondering how I can contribute to this ontology - maybe there are some specific tasks I could do? or is it just discussions for the moment? >>> >>> Best Regards, >>> Jeremy >>> >>> >>> [1] http://www.semantics.cc >>> [2] https://raw.githubusercontent.com/diachron/quality/master/src/main/resources/vocabularies/dqm/dqm.trig >>> >>> >> >
Received on Monday, 14 July 2014 19:08:06 UTC