On 12/08/2020 02:32, Irene Polikoff wrote:
For this to work, the spreadsheet will need to have a column with
these concatenated values.
As for the data in EDG, you can use SHACL rule to “infer” property
values - concatenating a company name with the department. While the
rule could be defined as a property value rule that infers a value on
demand, I don’t know if such dynamic inference would work for the
import matching logic. You could confirm it by trying, but I suspect
that for the match to work, you will need to execute the rules and
store the results prior to running the import. Use the Transform tab
to do this.
The out-of-the-box spreadsheet importer of EDG will not use inferred
values for matching (and this might be slow). As Irene said, you would
first need to materialize the inferences.
If you are somehow familiar with JavaScript then 6.4 offers more
flexibility for spreadsheet importing, including the ability to use
inferred values and building helper data structures for efficient
matching. See
http://www.datashapes.org/active/import.html
and in particular the demo video linked from that page
https://www.youtube.com/watch?v=Dn7O8siZpTc&feature=youtu.be
Holger
On Aug 11, 2020, at 12:05 PM, Fan Li <lifan0...@gmail.com
<mailto:lifan0...@gmail.com>> wrote:
Hi Irene. Thanks for the confirmation. I was thinking if I should
create a new property for Department and use SHACL rule to populate
it with concatenated company and department names. Then I can use
this new property for matching purpose only. Do you think if it would
work?
On Tuesday, August 11, 2020 at 11:04:20 AM UTC-4 Irene Polikoff wrote:
A property you use to find matches for related resources must
have unique across all instances of the related class (e.g.,
Department) in the asset collection you are importing into.
If there is no property that is unique, then property based match
will not work. You need to populate the Department column with
the actual URIs of these resources.
Hope this helps,
Irene
On Aug 11, 2020, at 10:57 AM, Fan Li <lifa...@gmail.com> wrote:
Suppose I am dealing with financial data related to two retail
companies "W" and "T" and each of them has a set of departments.
*"W":*
* Apparel
* *Electronics*
* Sports
*"T":*
* Clothes
* *Electronics*
* Sports & Recreation
As you see, some departments from different companies may have
the same name.
I have modeled the companies and the departments as instances of
Organizations and OrganizationalUnits and their relations
similar to the W3C Organization Ontology (the specific ontology
choice probably doesn't matter here).
Now I have the following financial data to import into EDG:
<Annotation 2020-08-11 105105.png>
I am using the "pattern-based" import strategy and trying to
match the company and department entities based on their label.
The challenge I am facing is that some department names are
ambiguous (e.g. "Electronics") and EDG would, understandably,
get it wrong sometimes.
I am asking the community what is the best practice to solve
this problem. Thank you!
--
You received this message because you are subscribed to the
Google Groups "TopBraid Suite Users" group.
To unsubscribe from this group and stop receiving emails from
it, send an email to topbraid-user...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/3aed4890-f945-4bbb-bb9d-3bcc52e551dcn%40googlegroups.com
<https://groups.google.com/d/msgid/topbraid-users/3aed4890-f945-4bbb-bb9d-3bcc52e551dcn%40googlegroups.com?utm_medium=email&utm_source=footer>.
<Annotation 2020-08-11 105105.png>
--
You received this message because you are subscribed to the Google
Groups "TopBraid Suite Users" group.
To unsubscribe from this group and stop receiving emails from it,
send an email to topbraid-users+unsubscr...@googlegroups.com
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/7a62cbf1-cfe2-4bf5-8d63-f4442af9e733n%40googlegroups.com
<https://groups.google.com/d/msgid/topbraid-users/7a62cbf1-cfe2-4bf5-8d63-f4442af9e733n%40googlegroups.com?utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed to the Google
Groups "TopBraid Suite Users" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to topbraid-users+unsubscr...@googlegroups.com
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/016FA346-C934-4B20-9EC4-89F0B91AD668%40topquadrant.com
<https://groups.google.com/d/msgid/topbraid-users/016FA346-C934-4B20-9EC4-89F0B91AD668%40topquadrant.com?utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed to the Google Groups "TopBraid
Suite Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to topbraid-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/8af698c7-52b9-3bfb-43a8-5562ea21d001%40topquadrant.com.
