Penn DB Group's logo
Visual Query Languages
Arrow; just used for page layout. People
Arrow, used for page layout Publications
Arrow, used for page layout Research
Arrow, used for page layout Classes
Arrow, used for page layout Seminar
Arrow, used for page layout Resources
   
Search this website

Visual Query Languages

Executive Summary

Complex data is of increasing interest, and arises naturally within a number of domains. In particular, scientific data is typically deeply nested and can be represented in a type system that allows arbitrary composition of record, variant, set, list and bag constructors. Schemas for these databases are large and often difficult to understand, even for a domain expert.

We have developed a query language called CPL for querying complex databases. The system -- called Kleisli -- has been used in the context of integrating biomedical databases, and has many benefits such as being easily extensible to encorporate new data sources, and an optimizer that improves queries spanning multiple data sources. Currently, we are working on an extended version of CPL/Kleisli, called K2. In contrast to CPL, this system has a richer type system and is able to model object identities and keys. It can be queried using OQL.

The VQL project is aimed at providing tools that appeal to the user's intuition and can be used to browse schema, and narrow down a large schema to a particular region of interest which then forms the basis for query generation through visual forms. The project has two components: The first is a schema browser and marking tool that allows the user to specify the region or regions of interest. The second component is a visual query language tool that translates form specifications into CPL. At any point, users may use the forms to browse the underlying instances and gain intuition about what the data represents. The tool will also be tied to BioWidgets, a visual display tool for biological data.

The project is completely implemented in Java using the Java Foundation Classes library. The following figure shows a schema describing two biological databases. The user can mark/unmark nodes of the tree and generate dialog boxes for querying such attributes.

Project Members

Susan Davidson   Hartmut Liefke   Julie Herman   Xilin Li   
Levine Hall
3330 Walnut Street
Philadelphia, PA 19104
 

Last update: 08/02/11     Comments