By integrating Machine Learning (ML) and Business Rules (BR) Rule Learner™ supports self-learning decision models capable to learn from its own results. A decision model that can learn from previous experiences and produce new knowledge become a good “employee” of the enterprise.
The above scheme shows the “ever-learning loop” when subject matter experts (business analysts) apply Rule Learner within the analytical world to produce the most appropriate business rules based on the previous decisions and their understanding of the changed business environment. In the operational world, newly produced rules will be used to generate new business decisions, which being saved in the Enterprise Data Repository, will serve as a source for further adjustments in the decision model. Watch Video
Analytical World. In the analytical world, business analysts take Historical Data from the Enterprise Data Repository and apply their expert knowledge to extract records (“Training Instances”) which can be used to learn classification rules. It can be done manually or using a rule engine (called “Rule Trainer“) that applies Training Rules created by subject matter experts to select only related training instances from the historical records. Rule Learner can execute different ML algorithms against the resulting training instances and analyze the generated business rules. This process could be repeated until subject matter experts believe they receive the adequate business rules ready for production. Then the generated business rules will be placed into the Enterprise Rules Repository to be executed by the latest version of the decision model/service.
Operational World. Being incorporated into business decision models, the automatically produced business rules will be used in the operational world to help to make business decisions using real-time data streams. New decisions will be produced by executing newly deployed decision model with the latest business rules. New data records, saved back into the Enterprise Data Repository, will serve as a source for further adjustments in the automatically generated business rules.
Success Stories. Rule Learner has been successfully applied to several customer projects, including the Internal Revenue Service (IRS) research “Automating Business Rules Creation Using Machine Learning Models“. Here is an excerpt from the IRS Performance Report: “IRS Management was satisfied with the services being performed by the Contractor and its staff members. The overall performance of Dr. Jacob Feldman and staff on behalf of The OpenRules, Inc., was very good. This company would be highly recommended to work for the Internal Revenue Service in the future in regards to Machine Learning (ML) Models Performance in IRS Enforcement Programs”. Later, the IRS awarded OpenRules with another contract, “Machine Learning Models Research and Development with NRP and LMSB Data“. More practical use cases with Rule Learner are described here.
Services. OpenRules, Inc. provides Technical Support for Rule Learner to help an enterprise jump-start the use of this integrated ML+BR solution. We also provide professional services to assist customers in specifying and implementing their rules discovery and building their “self-learning” decision models. Contact us at support@openrules.com.
Decision Intelligence Platform 