OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

set-comment message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Question about OASIS SET tools

I asked a question about an ambiguous XPaths matching of OASIS SET tools a month ago such as matching between (Supplier or Buyer)'s Primary_Identification.GLN_Identifier in GS1 and  (Supplier or Buyer)'s PartyIdentification.Identifier in UBL. I think the SET tool supports a kind of lexical mechanism to solve the issue. Could you let me know what kind of mechanism do the SET tools use for lexical analysis ? And, do you have any further progress to handle the issue ?

I am working for a XML schema mapping project at NIST, which basically provides a matching tool to analyze a lexical and structural information in the XML schema. If the current NIST project can improve the performance of the OASIS SET tools, I hope I can find a way to establish a collaboration project.

Thank you.
Best Regards,

Jaewook Kim

---------- Forwarded message ----------
From: Mustafa Yuksel <mustafa@srdc.metu.edu.tr>
Date: Mon, Sep 7, 2009 at 4:21 AM
Subject: {Disarmed} Re: Question about OASIS SET tools
To: "Kim, Jaewook" <jaewook.kim@nist.gov>
Cc: Erdem Alpay <erdem@srdc.com.tr>, "Prof. Dr. Asuman Dogac" <asuman@srdc.metu.edu.tr>, "Ivezic, Nenad" <nenad.ivezic@nist.gov>, "Prof. Yun Peng" <ypeng@umbc.edu>, senan@srdc.com.tr, set@lists.oasis-open.org

Dear Jaewook,

Considering the first issue, you are right. We handle this issue in the next step, where we present the matching XPaths for the corresponding ontology classes to the user. In this screen you will realize that, we match BuyerLocation with BuyerCustomer and SellerLocation with SellerSupplier through lexical mechanisms.

For the second issue, we will investigate more and let you know about the progress as soon as possible.

P.S. Please use OASIS SET TC channels for further communication related with ISU and other SET TC work. You may either become a member of the TC and use "set@lists.oasis-open.org" mailing list, or you can use the comment list. The latter is described in http://www.oasis-open.org/committees/comments/index.php?wg_abbrev=set:

"In order to verify your consent to the Feedback License terms and to minimize spam in the list archive, you must subscribe to the list before posting. To subscribe, send a blank email message to set-comment-subscribe@lists.oasis-open.org. Once you confirm your subscription, you may post messages (set-comment@lists.oasis-open.org) at any time. When posting to the comment list, please include specific information in the "Subject: " line identifying the topic, especially in communications that provide input on TC specifications."

I am copying this email to SET TC mailing list.

Thank you and best regards,


Jaewook Kim wrote:
Dear Dr. Asuman and Erdem,

I could install your software and get a mapping result. Thank you for your help.
I have a few questions and issues about the solution as follows:

First question is: When the software finds a multiple mapping, it shows that "x GS1 and y UBL matches found". Can it also find which pair of elements are mapped among x GS1 and y UBL elements ?

For example, in the mapping result between GS1's and UBL's "Trade Item Location Profile" messages, the software finds a equality mapping between "PartyIdentification.Primary_Identification.GLN_Identifier" a element of GS1 and "PartyIdentification.Identifier" a element of UBL. And it shows the detail result such as "4 GS1 and 2 UBL matches found", but doesn't show that which pair of elements among 4 GS1 and 2 UBL elements can be mapped. 

The found 4 GS1 elements' node paths are as follows:

1. Trade Item Management Profile.Details -> Trade Item Management Profile.Profile.Trade Item Management -> ... -> Collaborative Item.Buyer Location.Party -> Party Identification.Details -> PartyIdentification.Primary_Identification.GLN_Identifier

2. Trade Item Management Profile.Details -> Trade Item Management Profile.Profile.Trade Item Management -> ... -> Collaborative Item.Seller Location.Party -> Party Identification.Details -> PartyIdentification.Primary_Identification.GLN_Identifier

3. Trade Item Management Profile.Details -> Trade Item Management Profile.Inheritance_Association ->  ... -> Plan Document.Buyer.Party Identification -> Party Identification.Details -> PartyIdentification.Primary_Identification.GLN_Identifier

4. Trade Item Management Profile.Details -> Trade Item Management Profile.Inheritance_Association ->  ... -> Plan Document.Seller.Party Identification -> Party Identification.Details -> PartyIdentification.Primary_Identification.GLN_Identifier

Also, 2 UBL elements' node paths are

1. Trade Item Location Profile.Details -> ... -> Information Request Trade Item.Buyer_Custmer -> Customer Party.Details -> ... -> PartyIdentification.Identifier

2. Trade Item Location Profile.Details -> ... -> Information Request Trade Item.Seller_Supplier -> Supplier Party.Details -> ... -> PartyIdentification.Identifier

I think the first and third elements in GS1 can be mapped to first in UBL and second and third in GS1 can be mapped to second in UBL. Can the iSURF aproach identify these kinds of mapping ? I am thinking how to improve iSURF approach by finding these kinds of mapping.

The second issue is
Some mapping items retrieved by "Load Equality List" are in neither GS1 nor UBL schemas. For example, in the mapping of GS1's and UBL's Focast messages, I couldn't find any mapping item among the retrieved equality list. When I clicked "Find Equivalences" in the right button menu, it shows a error such as "Index '0' specified is out of bounds".

Thank you.
Best Regards,

Jaewook Kim

On Thu, Aug 6, 2009 at 8:31 AM, Erdem Alpay <erdem@srdc.com.tr> wrote:
Hi Nenad,

OASISSET.zip file which is in
( http://www.srdc.metu.edu.tr/iSURF/OASIS-SET-TC/tools/OASISSET.zip )
is now obsolete. If you want to setup  yourself, you can find the updated
version in the following link:


The installation guide is inside the "docs" folder in the zip file.

If you prefer to use the tool interface directly, it is already deployed
on one
of our servers. You can reach it from the following link:

MailScanner warning: numerical links are often malicious:

You can find information about using this interface in the iSurf
that Prof. Dogac previously sent to you.

I am at your disposal for any further information you might need.

Best regards,


Nenad Ivezic wrote:
> Hi Asuman:
> Thanks for the pointers.
> Last time we talked about us here taking look at SET, you mentioned
> that you were still working on the tools.
> Is the link below to the OASISSET.zip file the right place from which
> to install the tools?
> As soon as Kim returns, I would like to start looking into the SET
> deliverables.
> Best regards,
> - Nenad
> Prof. Dr. Asuman Dogac wrote:
>> Hi Kim,
>> You will find the information you need in the following
>> iSURF Deliverable.
>> Best regards,
>> Asuman
>> Kim, Jaewook wrote:
>>> Dear Dr. Asuman,
>>> I appreciate your help.
>>> I couldn't find the SET TC Deliverable from OASIS SET TC site, but
>>> could download the tool using the following link:
>>> http://www.srdc.metu.edu.tr/iSURF/OASIS-SET-TC/tools/OASISSET.zip
>>> And, do you have the UBL and GS1 schemas that SET tools used for the
>>> mapping ? I think the schemas will be very useful to review the SET
>>> approach and review my research.
>>> I will be out of office from now to Aug 13th, so I may not be able
>>> to check your message.
>>> Thank you.
>>> Best Regards,
>>> ________________________________________
>>> From: Prof. Dr. Asuman Dogac [asuman@srdc.metu.edu.tr]
>>> Sent: Monday, August 03, 2009 11:31 AM
>>> To: Kim, Jaewook
>>> Cc: Ivezic, Nenad; Prof. Yun Peng; erdem@srdc.com.tr;
>>> senan@srdc.com.tr; suat@srdc.com.tr
>>> Subject: Re: Question about OASIS SET tools
>>> Dear Kim,
>>> Thank you for your email and your interest in SET TC work.
>>> Indeed, SET Harmonized ontology discovers the equivalences
>>> of structurally similar elements; not only BBIEs but also ABIEs.
>>> For structurally different elements we resort to providing rules.
>>> Additionally, in finding the similar BBIEs we also have a heuristic
>>> rule
>>> equating 3 of the 14 data types because that is the way
>>> they are used in real life. We use the similarity of "Property Terms"
>>> to rank the results.
>>> You can download the tools and try them by yourself.
>>> For the tools, Erdem, Suat or Senan can provide you help.
>>> All the information is available in the SET TC Deliverable
>>> at the SET TC site. If you have difficulty in downloading
>>> it let me know I will send you.
>>> Enclosed please also find a paper describing our work.
>>> Once again thank you and best regards,
>>> Asuman
>>> Jaewook Kim wrote:
>>>> Dear Dr. Asuman,
>>>> I am a guest researcher at NIST working with Nenad. Nenad gave me a
>>>> slide about OASIS SET-TC approach because it seems like a related
>>>> to my research which is to discover mappings between two XML schemas
>>>> (without upper-ontology). I reviewed the slide and have a few
>>>> questions as follows:
>>>> 1. Can the SET Harmonized Ontology approach discover mappings between
>>>> the structured artifacts ? (e.g., mapping between ABIE and ABIE) I
>>>> think the SET Harmonized Ontology is only useful to discover the
>>>> mappings between atomic elements which are the leaf of the schema
>>>> tree, so the SET Heuristic Rules are used to discover mappings
>>>> the structured artifacts, am I right ?
>>>> 2. If the schema contains multiple BBIEs in different location
>>>> look at the attached slide), can the SET Heuristic Rules discover the
>>>> mapping correctly ? In the attached slide, the gray lines are
>>>> discovered mapping by SET Harmonized Ontology and red lines are
>>>> correct mapping.
>>>> I am investigating if my research can help the schema mapping problem
>>>> that SET approach is dealing with. I hope I can get an additional
>>>> information.
>>>> Thank you.
>>>> Best Regards,
>>>> Jaewook Kim
>>>> ---------- Forwarded message ----------
>>>> From: Nenad Ivezic <nivezic@nist.gov>
>>>> Date: Wed, Jul 22, 2009 at 2:29 PM
>>>> Subject: A potential problem for your PhD work
>>>> To: Jaewook Kim <jaewook@nist.gov>
>>>> Kim:
>>>> I promised to send you material to provide a problem that may be a
>>>> basis for your PhD work.
>>>> Take a look at

>>>> and in particular slide 39 and let me know if this problem in your
>>>> view could be attacked by your methodology.
>>>> Let me know what you think.
>>>> - Nenad
>>>> ps.  This is also effort headed by our friends in Turkey and I
>>>> they would be interested to share data.
>>> PS. Please note the change in the phone and fax numbers. Thank you.
>>> --
>>> Professor Asuman Dogac             email: asuman@srdc.metu.edu.tr
>>> WWW: http://www.srdc.metu.edu.tr/~asuman/
>>> Director                           Phone: +90 (312) 210 1393, or
>>> Software R&D Center                       +90 (312) 210 1763
>>> Department of Computer Eng.        Fax: +90 (312) 210 1837
>>> Middle East Technical University
>>> 06531 Ankara Turkey                      skype: adogac
>> PS. Please note the change in the phone and fax numbers. Thank you.

Mustafa Yuksel

Software Research and Development Center
Department of Computer Eng.
Middle East Technical University
06531 Ankara TURKEY

Phone: +90 (312) 2102076
Fax: +90 (312) 2105572

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]