| The growth of information in corporate databases has resulted in a new field of data management called “data warehousing”. In a data warehouse, data is extracted from multiple sources, transformed to be consistent among those sources and presented to the user for ad hoc querying, analysis of trends or other decision support purposes.; The goal of the user is to extract the data efficiently. That is not always easy without an understanding of the underlying structure of the database or knowledge of SQL. Users need to understand the basic concepts of relational databases in order to extract this data, which can be a challenging task for novice users. This thesis proposes an architecture and implements a tool that guides the user through the extraction process and helps the user find attributes of interest in a step by step manner without requiring an understanding of the underlying physical database environment. The user defines an agent that monitor's the data warehouse for thresholds and alerts the user when these thresholds are crossed. The architecture consists of three autonomous inter-dependent sub-systems consisting of a client Query Builder, a server Agent Coordinator and an Execute, Monitor and Deliver agent. The benefits are a scalable, extensible and maintainable architecture for ad hoc querying and data mining that is usable by novice computer users. |
