GOING BEYOND DATA MINING

Services

Products

CEM

Climate Change

Energy and
    Environmental Decision
    Support Systems

Unique Technologies

 Papers:

Beyond Data Mining

Process Management

Distributed/Approximate
     Reasoning

Knowledge gained from data mining has allowed many organizations to increase productivity, cut costs, gain insight into business processes, or make better decisions in complex situations. However, data mining is only the first step on the road to more effective business operations. Although "mined" knowledge is valuable it is limited for a number of reasons. REDUCT Decision Suite (DataLogic, RuleDesigner and DecisionExpert (Forecaster and Optim)) addresses the major deficiencies and limitations of data mining methods as described below:

1. Your data are not a result of exact measurement and that which is measured is often ill defined.

Data are linked to representation of the system and may be simplified or aggregated. To overcome this problem, DataLogic uses new forms of data intelligence based on an extension of the set theory. The method provides a theoretical framework for data analysis and knowledge discovery from ambiguous data. It is based on the premise that generalization of information helps to uncover regularities in data. DataLogic software provides rigorous mathematical techniques for generation of information from data, and for analysis of knowledge contained in databases.

2. The future is not totally represented by the past.

The future almost always conceals something unpredictable or indeterminable. This is because complex systems cannot be determined in a comprehensive way. Thus, in addition to the external factors (environment) responsible for the uncertainty, there are internal factors which become prominent under different conditions. RuleDesigner addresses this problem by using distributed reasoning methods. These methods employ distributed, cooperative intelligent agents to generate new objects (ideas/knowledge) not present in your database. This knowledge can be then applied in decision making.

3. Your model is not the system but only a representation of the system.  In addition to omissions, simplifications and aggregations, there are other factors which have not been captured by the model. DecisionExpert uses approximate reasoning methods to compensate for poor definition of the model (e.g. attribute values, outcomes, consequences). DecisionExpert handles uncertainty, conflicting constraints and information, and provides more consistent, higher quality decisions.

For more information, please contact us.

REDUCT & Lobbe Technologies Inc.
P.O. Box 800,  186 - 8120 No.2 Road., Richmond, BC,  Canada  V7C 5J8
ph: (604) 275-3711   fax: (604) 275-3711  email: dispatch@reduct.com