Add object properties "has aggregation type" and "has time stamp alignment"

[DevPy129387]

Description of the issue

We would like to define / specify a time series with OEO. Therefore we define a "data set", which has a series of "data items". Now we have to to specify for every "data item" (or better for the whole dataset) an "aggregation type" and a "time stamp alignment". But I can't find appropriate object properties to set these properties.

(Is it possible to specify "unit", "aggregation type" and "time stamp alignment" only once for the whole "data set"?)

Ideas of solution

Add the object properties "has aggregation type" and "has time stamp alignment" to the oeo.

Workflow checklist

• [x] I discussed the issue with someone else than me before working on a solution
• [x] I already read the latest version of the workflow for this repository
• [x] The goal of this ontology is clear to me

I am aware that

• [x] every entry in the ontology should have a definition
• [x] classes should arise from concepts rather than from words

[stap-m]

Thanks for the issue @DevPy129387. It would be reasonable to add the requested relations to OEO.

For the "has aggregation type" I suggest the follonging:

• We have already An aggregation type is a data descriptor that contains information on the aggregation method applied on a data set.
• Add new object property has aggregation type: A relation between a data set and an aggregation type that indicates the method of data aggregation applied to the data set.

EDIT: Maybe we should apply the aggregation types even to data item: data set is a subclass thereof and thus, it would be applicable to both: signe items or data sets.??

[stap-m]

Here some first ideas for "has time stamp alignment":

• For time stamp we have already: A time stamp is a data descriptor that is used to describe a zero-dimensional temporal region, or a data item that is related to a zero-dimensional temporal region, e.g. a time stamped measurement datum. However, I think, this definition is not sufficient in the context of time stepts / time series.
• I suggest the following:
  • Update definition of time stamp: A time stamp is a data descriptor that is used to describe a temporal region, or a data item that is related to a temporal region, e.g. a time stamped measurement datum. If the time stamp refers to a time step/one-dimensional temporal region, the indication of a time stamp alignment is required. [^1]
  • Add object property has time stamp alignment: A relation between a time step or time series[^2] and a time stamp alignment that indicates the position of the time stamps that refer to the time step(s).

[^1]: A time step is a one-dimensional temporal region that has a start time and an ending time and thus a finite duration. [^2]:A time series is a data set that references to a set of time steps or zero-dimensional temporal regions.

[DevPy129387]

These suggestion sound very good! To have the aggregation type for both - data items and data sets would be good.

[stap-m]

Any feedback from oeo devs?

[l-emele]

• Add data property has time stamp alignment: A relation between a time step or time series2: and a time stamp alignment that indicates the position of the time stamps that refer to the time step(s).

The two classes time step and time series are of two very different types: the first is a temporal region and thus an occurent while the latter is a data item and thus a generically dependent continunant). I don't think that it is as good and future-proof solution to implement one object property that has two so very different concepts as their range. Instead we should implement here two distinct object properties.

[stap-m]

• Add data property has time stamp alignment: A relation between a time step or time series2: and a time stamp alignment that indicates the position of the time stamps that refer to the time step(s).

The two classes time step and time series are of two very different types: the first is a temporal region and thus an occurent while the latter is a data item and thus a generically dependent continunant). I don't think that it is as good and future-proof solution to implement one object property that has two so very different concepts as their range. Instead we should implement here two distinct object properties.

That's a good point. I think, the best solution would be to only refer to the entities describing the temporal regions (GDC), not the temporal regions itself. In our case, probably time stamp instead of time step.

[stap-m]

Any more comments or can we implement?

[stap-m]

@madbkr oeo-model seems also the right place to me