Skip to main content
AMIA Annual Symposium Proceedings logoLink to AMIA Annual Symposium Proceedings
. 2005;2005:1019.

Automatic XQuery Generation and Generalized Visualization for an XML Interface to a Relational Database

E Sally Lee 2, Dan Suciu 3, James F Brinkley 1,2,3
PMCID: PMC1560423  PMID: 16779306

Abstract

XBrain is an application that facilitates data exchange by dynamically publishing relational data over the web in XML format. In the current work we further enhanced its functionality through automatic query generation to aid the human users. We also extended its visualization tools by developing them into generalized XML visualization web services.


XBrain is a JSP web application that facilitates data exchange by automatically generating XML from a relational database containing language map data1. XBrain takes XQuery as input, and interacts with a relational database through SilkRoute, an application that dynamically converts XQuery to SQL (Figure 2). XML results are then returned to the user as HTML, CSV, or a dynamically generated image. Although the users (neuroscientists) found the results of XBrain to be useful, they found XQuery too difficult to formulate. Furthermore, the previously implemented XML visualization tools were too specific to the XBrain application to be reusable.

Figure 2.

Figure 2

XBrain Architecture

From interviews, we found that users wanted a more intuitive query interface that automatically generates XQuery. To this end, we developed a graphical query interface front end to XBrain (Figure 1).

Figure 1.

Figure 1

Graphical User Query Interface

With this interface, which is written in JSP, users can visually select and filter on a set of common variables. The application automatically generates an XQuery, based on the user specified criteria and a pre-defined XQuery template. The template structure was determined by assessing what are the most common types of query asked by the users. The users found this query interface much easier to use. The new query interface also aided the users in analyzing the underlying data more rapidly.

In order to make the XML visualization tools more general, we transformed them into individual XML visualization web services. Each HTML, CSV, and dynamic image web service takes an XML file as input and returns visualization as output (Figure 2). The HTML and CSV web services can be used by any application to visualize any XML. The dynamic image web service can only be used to visualize structural brain data, but there are many applications that generate such data.

Acknowledgements

NIH HBP Grant DC 02310, NLM Training Grant T15 LM07442

References

  • 1.Tang, Z. and Kadiyska, Y. and Li, H. and Suciu, D. and Brinkley, J. F. (2003) Dynamic XML Based Exchange of Relational Data: Application to the Human Brain Project. In Proceedings, American Medical Informatics Association Fall Symposium, pages 649–653 [PMC free article] [PubMed]

Articles from AMIA Annual Symposium Proceedings are provided here courtesy of American Medical Informatics Association

RESOURCES