Goal: Reverse Engineering A Target Database
An insurance company was to take over the nursing home and health care business of another insurance company. The database to be acquired was generally undocumented and may have contained errors. Before completing the deal, it was necessary to perform due diligence and establish the legitimacy of the database and have an understanding of how it was created.
WizRule was applied to the database. First, inconsistencies and errors were detected. For example, in dental files, gender was classified three ways – male, female and others.
Then WizRule analyzed the database to reverse engineer it. Since documentation was limited, the company had no way of determining how the database had been created. This was very important, since the company would now have to maintain and add to the database. WizRule revealed all the rules and relationships which existed in the database and thereby uncovered the original structural rules. Among the unusual patterns discovered was that a short-cut medical coding had been used which classified all patients, both male and female, with certain characteristics as having a particular type of disease known to be almost exclusively female in nature.
The legitimacy of the database was established. The data was cleansed and the logic of the database was revealed. The deal was completed.
