MOEngine

Add state-of-the-art rule discovery to your software.  The heart of the state-of-the-art Magnum Opus rule discovery system, MOEngine is available for your use as a Windows API, Unix Library, or C++ source code.

MOEngine brings to your software the power of the most recent advances in rule discovery technology.  It delivers unrivaled speed and flexibility.

MOEngine is designed to handle large data sets containing up to millions of records, although its capacity to do so will be constrained by the amount of memory available.   It empowers the host to select between five measures of the importance of a rule, leverage, lift, strength, coverage, or support, providing alternative analyses of the same data.

Most association rule software is designed for processing transaction data and relies on the data being sparse (most transactions contain very few of the available values). In contrast, MOEngine does not rely on sparse data for efficient processing. As a result, it is useful for mining a wide variety of data.

MOEngine allows the host to specify the maximum number of rules to be found, and place restrictions on rules to be considered.   Within these restrictions, it finds the rules with the highest values on the specified measure of importance. It will only find fewer than the specified number of rules if the search is terminated by the host or there are fewer than the specified number of rules that satisfy the host specified constraints.

Some important features:

• MOEngine is easy to use and does not presume advanced knowledge of Statistics or Machine Learning.

• MOEngine allows you to customize the presentation of rules to a format that suits your application.

• MOEngine can automatically filter out rules that are unlikely to be of interest.

• MOEngine finds rules from both basket data and attribute-value data.

• MOEngine has been designed to analyse substantial databases containing up to millions of records (although its capacity to do so may be limited by the amount of memory installed in the host computer).

• Unlike other rule discovery software, MOEngine does not rely on sparse data for efficient processing.

• MOEngine is fast. It has linear compute time. If the amount of data is doubled then the compute time will approximately double (unless the available physical memory is exceeded - use of virtual memory may increase compute time considerably).

Email us for further information on how easy it is to incorporate MOEngine into your software.