|
Picalo started in 2000 when Conan C. Albrecht, a professor at Brigham Young University, compiled his fraud detection routines into a single platform and released it as an open source platform. He had been writing analysis programs throughout the 1990's and wanted a more integrated toolset. Following is a quote from him:
For years, I used many different applications for analysis -- Excel, Paradox for DOS, Visual Basic, Microsoft Access, ACL, IDEA, PostgreSQL, Numeric Python, SAS, and many others. Each of these applications had strengths, but none was specifically written for analysis of corporate data or for data mining for fraud schemes. Some were generic databases; others were audit-oriented or statistical applications. When I went searching for an analysis application, I had the following requirements:
- Cross platform so analyses could run on Windows and Mac and Linux laptops as well as corporate Unix servers.
- Routines specific to the detection of fraud and corruption.
- Support for routine automation in a robust, powerful (yet easy to use) language.
- Ability to analyze huge amounts of data.
- A way for technical analysts to support non-technical analysts without the typical "have the techie do all the hard work" result.
- Both a graphical user interface and bare-bones, console interface.
Finding no product that suited my needs, I developed Picalo, an open architecture that all can contribute to and use.
In 2005, Picalo gained a user interface as the United Nations and World Bank wanted to use the platform to detect internal fraud. This website was created that year to provide a community home for Picalo.
The primary developer of Picalo is still Conan Albrecht. He has received help from research assistants as well as input from many individuals in the field.
We are actively looking for additional developers of the Picalo platform. As it progresses, we hope it will provide the basis for a worldwide repository of fraud detection routines.
|
