ADAPA

:"This article describes ADAPA, a decision engine used to manage and design automated decisions systems. For the Babylonian and Summerian god of wisdom and of the ancient city of Eridu see Adapa."

ADAPA is intrinsically a decision engine. It combines the power of predictive analytics and business rules to facilitate the tasks of managing and designing automated decisions systems.

Automated decisions

When first released, ADAPA (Adaptive Decision And Predictive Analytics) was purely a scoring engine, used to produce scores out of statistical models expressed in PMML (Predictive Model Markup Language) format. With the addition of a rules engine to its core, ADAPA is now able to seamlessly combine rules and predictive models, which enables businesses to manage and design automated decision systems. In this way, ADAPA allows for the concretization of Enterprise Decision Management (EDM) solutions.

PMML support and predictive analytics

Predictive analytics comprises a series of modeling techniques which can be used to extract relevant patterns present in large amounts of data to better predict the future.

ADAPA is able to generate scores out of a variety of predictive modeling techniques expressed in PMML. PMML provides a standard way for the expression of predictive models. In this way, proprietary issues and incompatibilities are no longer a barrier to the exchange of models between applications.

Currently, ADAPA supports the following PMML elements:

* Decision Trees
* Neural Networks
* Support Vector Machines
* Linear and Logistic regression as well as all general regression PMML models:
** Multinomial Logistic
** General Linear
** Ordinal Multinomial
** Simple Regression
** Generalized linear model

Once a model is uploaded in ADAPA, it can be executed in batch and real-time. ADAPA is a PMML consumer, therefore it is able to execute PMML code exported from tools such as R, SPSS, IBM, etc.

Besides offering a web-based console to manage models, ADAPA includes capabilities to test models for accuracy of scoring.

Business rules

Business rules allow for business process and logic to be expressed outside of programming code.

ADAPA allows for business knowledge to be expressed in simple tabular format. In ADAPA, rules can be used to manage the execution of different predictive models depending on the business context. They can also incorporate scores generated by different predictive models throughout the business process. The acting together of the two technologies has the potential to significantly extend the precision of any decision logic.

ADAPA rules leverage the power of the leading Java open-source rules engine Drools which is supported by a strong community of developers and JBoss, a division of Red Hat. This fast, highly efficient rules engine has proven its excellence in numerous commercial installations.

All decisions in ADAPA are readily available by the use of Web Services which follow the Java Data Mining (JDM) specification.

ADAPA To Go

ADAPA has recently been released as an iGoogle gadget which allows for anyone anywhere to access ADAPA predictive analytics functionality for free. A model is uploaded in PMML 3.2 and made available for verification or scoring right away. Zementis (the maker of ADAPA) also released a PMML converter as an iGoogle gadget. This allows for users to convert older PMML models (versions 2.1, 3.0, and 3.1) to version 3.2.

A paid version of ADAPA predictive analytics is now available through the Amazon Elastic Compute Cloud (Amazon EC2). This version provides the first SaaS (Software as a Service) predictive decision engine. Unlike the free version, the user here can upload and manage several models and score data in real-time through the use of web-service calls. ADAPA as a Service empowers people, since it allows for anyone anywhere to deploy and use state of the art data mining models.

ADAPA Editions

ADAPA is currently being offered in two editions:

* Predictive Analytics Edition: contains the predictive analytics engine as well as web-services functionality.
* Enterprise Edition: in addition to the predictive analytics engine and web-services, it also includes business rules functionality and a reporting engine. This edition allows for the seamless interplay of rules and predictive models.

External links

* [http://smartenoughsystems.com/wp/2008/06/20/buying-predictive-analytics-like-books-zementis-adapa/ Buying predictive analytics like books] - Smart Enough Systems blog on the ADAPA Amazon EC2 offering.
* [http://www.google.com/ig/adde?hl=en&moduleurl=hosting.gmodules.com/ig/gadgets/file/115640297026242314759/adapawidget.xml ADAPA iGoogle gadget] - gadget that can be used to access ADAPA predictive analytics from your own console.
* [http://www.google.com/ig/adde?hl=en&moduleurl=hosting.gmodules.com/ig/gadgets/file/115640297026242314759/converterwidget.xml PMML Converter iGoogle gadget] - gadget to convert older versions of PMML to PMML 3.2.
* [http://adapasupport.zementis.com ADAPA support] - blogs providing support for the ADAPA demo and the PMML converter gadgets as well as the paid version of ADAPA - Amazon EC2.
* [http://www.zementis.com/adapa.htm ADAPA product page] - contains information about ADAPA as well as all the open standards it supports.
* [http://www.zementis.com/pmml_exporters.htm PMML Exporters]
* [http://www.dmg.org/ Data Mining Group (DMG) ] - describes PMML, the Predictive Modeling Markup Language, as well as gives information on all the companies currently supporting the standard.
* [http://labs.jboss.com/drools/ Drools homepage]