Transformation Rule Discovery through Data Mining

Data-intensive software programs typically transform a set of source data values into target data values. While the group of developers who write, compile and test the software or the query have a clear understanding of the transformation logic (or transformation rules) at the time the program is created, that understanding can quickly fade at an enterprise level for a variety of reasons, including poor documentation, loss of the source (uncompiled) version of the software, loss of the developers who wrote the software, or lack of available skills in the programming language (e.g., COBOL). This leaves the enterprise in a precarious position of not being able to maintain, upgrade or migrate the software programs at the heart of their operations unless they can recreate the transformations that relate the source data to the target data. In this paper, we propose a technique to reverse engineer transformation rules by analyzing the data using data mining algorithms and processing the results of these algorithms. We demonstrate our technique using a prototype implementation and prove its correctness with a sample data set.