Figure 1 - uploaded by Michael Saelee
Content may be subject to copyright.
Typical Internet data integration architecture 

Typical Internet data integration architecture 

Source publication
Conference Paper
Full-text available
We describe a mediator designed specifically for an intranet rather than an Internet. Our intranet mediator allows data reconciliation and integration to be completed long before the query is executed. This is because information about the data sources is known and a combined schema can be defined prior to the query. On the Internet, source schemas...

Context in source publication

Context 1
... typical architecture for internet data integration sys- tems is shown in Figure 1. The mediation unit in internet data integration systems is complex, as much work needs to be done during query time to ensure schema integration. ...

Similar publications

Conference Paper
Full-text available
We describe a methodology for mapping linear recurrence equations to a spectrum of systolic architectures. First, we design a systolic program in a very general architecture referred to as Basic Systolic Architecture and establish the correctness of the implementation. Next, we show how efficient transformations/implementations of programs for diff...

Citations

... Much of our work has focussed on the integration of structured data with text [14] and has been implemented and deployed by both government and commercial organizations of varied sizes. Our recent work on intranet mediators takes data from a data warehouse and seamlessly integrates them with less structured data yielding a unified view of data regardless of where or how they are stored [20]. In addition to unified access, we have worked on using information fusion to improve effectiveness [6]. ...
Article
Full-text available
We describe our progress extending the undergraduate Computer Science (CS) curriculum to include a deep understanding of information retrieval (IR) and data mining (DM). Instead of simply understanding how to build applications using tools involving IR and DM, students build these tools and learn the relevant algorithms implemented in these tools. Some novel approaches exist in our work. We include a hands-on lab setting where students use the tools they have built to perform experiments that could ultimately extend the field. Hence, undergraduates have firsthand knowledge of performing research in Computer Science using a scientific method. Secondly, we have a rigorous set of evaluation criteria developed by our Psychology department that will evaluate how well students learn using our novel approaches. Ultimately, we believe these two courses warrant consideration into standards developed for the undergraduate CS curriculum.