WebSmatch Project Page

ZENITH Team

 


WebSmatch (Web Schema Matching) is a flexible, open environment for discovering and matching complex schemas from many heterogeneous data sources over the Web. It provides three basic functions: (1) metadata extraction from data sources; (2) schema matching (both 2-way and n-way schema matching), (3) schema clustering to group similar schemas together. WebSmatch is being delivered through Web services, to be used directly by data integrators or other tools, with RIA clients.

Project members :

  • Scientific manager : Patrick Valduriez
  • Project manager : Rémi Coletta
  • Software engineer : Emmanuel Castanier
  • Contributors : Zohra Bellahsene, Hoa Ngo Duy

    • WebSmatch architecture :

    Schema matching is a fundamental and very difficult task of data integration. Based on our experience gained in developing original schema matching tools and industrial-strenght data integration systems, we are developing the WebSmatch software, a flexible, open environment for Web schema matching. To foster its use by data integrators over the Web, WebSMatch has a modern architecture based on Web services and RIA (Rich Internet Application). A WebSmatch client is an RIA to be used within any modern browser, thus providing an easy, yet powerful GUI. Our RIA is implemented in GWT (Google Web Toolkit). A client can call WebSmatch services through Web services to perform various functions such as

  • extracting metadata from data sources, e.g. xls files, using machine learning techniques;
  • matching two or more schemas, using different matchers which are made independent of input format (through the use of a metametamodel);
  • clustering similar schemas;
  • importing/exporting in any format.

    • Each service can be used independently by any kind of client that supports standard web service calls.



    Demo

    A beta online tool is available to demonstrate how it works. With this tool, you can currently:

  • import files (OWL, RDF, RDFS, XLS supported);
  • visualize schemas using tree or cell representation;
  • match two schemas, using different matchers;
  • see the results as a list of scores;
  • generate a cluster view.



    • To try it, go to

    WebSmatch beta tool

     Requirements :
  • modern web browser supporting JavaScript
  • 1024x768 pixels resolution or better

    • Videos :

    End to end demonstration video from metadata detection to visualization (using Google DSPL format).

    WebSmatch supports the full process of importing, refining and integrating data sources and uses third party tools for high quality visualization. We use a typical scenario of public data integration which involves problems not solved by currents tools: poorly structured input data sources (XLS files) and rich visualization of integrated data.

    The first video will demonstrate the export/visualize feature from WebSmatch to DSPL visualizator.

    Loading video ...

    The second video will demonstrate the metadata integration feature with visualization of the produced document. You can download the 2 XLS files used in this demonstation :

    Gaz emissions

    and

    World temperature evolution

     You can try to reproduce it with the account "demo" and pass "demo".

    Loading video ...

    How to use the tool

    First you will have to login. Just use the "test" login and password



    You will face the main interface. Just click on the menu bar "File" and "Load schema from store".



    Now you will face a list of already imported schemas, just choose one and click to go to visual representation.



    After loading a schema imported from an XLS file, you will see the editor frame. You can modify the schema (yellow cells are attributes), when satisfied, just click on "Store XLS file as a shema", you will have to choose a name for it and describe it. The description must be as precise as you can as it will be used for the matching in addition of other informations.



    You can now import a new schema.



    After integrating and validating a new schema, you can go to the matching frame by clicking on the menu bar.



    Then, you get this frame :



    Just choose left and right schemas and then match them. You will go to the results frame. At this point, adjust the threshold and click on "Show results".



    You can also use the cluster view, click on "View results/As a cluster".



    Here is an exemple of clustering :



    You can also use the two way view to edit the matches, click on "View results/Two way view". To add an expert match, click on a node, then on a second node, it will create the expert match.




    (c) INRIA - LIRMM - ZENITH Team 2010

    Powered By GForge Collaborative Development Environment