IRS: Automating Business Rules Creation Using Machine Learning Models

Washington, DC and Edison, NJ -- November 13, 2007 -- OpenRules, Inc. has successfully completed the government contract “Automating Business Rules Creation Using Machine Learning Models” issued by the National Headquarters Office of Research, Internal Revenue Service (IRS).  The objective of this project was to integrate machine learning results into a standard business rules environment in a way that allows IRS subject matter experts to easily identify any gaps between the rules developed by its experts to detect potential non-compliance and the rules automatically developed by the machine learning system. 

To meet this objective, OpenRules has applied the Rule Learner, a machine-learning component of its popular open source business rules management system, to the National Research Program data provided by NHQ Research.  As a result, the Rule Learner was able to successfully generate business rules that model the issue selection domain.  OpenRules applied different ML methods to a large volume of historical data and generated the learned rules in a form automatically executable by the OpenRules engine.  The achieved classification results are compared well with results previously achieved by manual issue selection.   The resulting approach demonstrated a generic, yet fully automated integration of the learned rules into a Business Rules Management System. 

IRS does not endorse any vendor or product and, in fact, one of the critical objectives of this research was to construct a methodology whereby the relationships discovered by a machine learning component could be converted into rules and run on the enterprise Rule-Based Management System regardless of the Rule Engine in current use.  According to IRS Project Manager, “we wanted to automatically discover rules that were meaningful to our business analysts and at the same time could be run on the Enterprise BRMS.  We accomplished all of these objectives.”  An IRS domain expert reviewed the automatically learned rules and suggested them to be “useful, comprehensible, and intuitive.”  While generated rules do not necessarily replace manually elicited rules, they are extremely valuable for analysis and comparison.  Generating rules will improve IRS capabilities to keep up with a constantly changing environment.

The IRS expects that an integrated approach of automated knowledge discovery with business rules authoring could potentially be useful as a sophisticated feedback mechanism for operational enhancements.

About Internal Revenue Service
The information about IRS can be found at http://www.irs.gov

About OpenRules
OpenRules, Inc. is a NJ-based government contractor whose specialists are known to many leading US businesses as experts in developing practical decision support applications using different decision support technologies.  OpenRules, Inc. is the company that develops, supports, and markets the OpenRules™ product and provides related consulting and training services.  For more details, please visit http://www.openrules.com/.