Nautilus Systems, Inc. logo and menu bar Site Index Home
News Books
Button Bar Menu- Choices also at bottom of page About Nautilus Services Partners Case Studies Contact Us
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [Subscribe]

DM: PolyAnalyst COM - Toolkit of DCOM-based Machine Learning Components

From: Sergei Ananyan
Date: Thu Nov 11 21:29:01 1999

PolyAnalyst COM ­ Toolkit of DCOM-based Machine Learning Components

November 11, 1999
Megaputer Intelligence

In maintaining a tradition of innovation in the development of data mining solutions, Megaputer Intelligence proudly announces the availability of PolyAnalyst COM ­ a comprehensive toolkit of data mining modules based on Microsoft’s Distributed Component Object Model (DCOM) technology. PolyAnalyst COM provides individual machine learning algorithms for their effortless incorporation into external decision support applications.

Megaputer invites integrators and developers of various decision support systems to take advantage of the power and convenience of these affordable DM components. PolyAnalyst COM modules can be easily integrated into developed applications with the help of simple development tools like Visual Basic for Applications (VBA). The DCOM architecture makes created vertical applications easily extendable, upgradable and customizable, while allowing users to exploit the unique analytical capabilities of PolyAnalyst through an already familiar user interface. In addition, the new architecture provides an elegant solution to the problems of data import and scoring, since the data and the model are now stored in a single native environment. PolyAnalyst COM fully supports Client/Server architecture.

In addition to advanced data import, manipulation, visualization, scoring, and report generating capabilities, PolyAnalyst COM offers the following machine learning algorithms as individually available DCOM modules:

1. PolyNet Predictor (GMDH-Neural Network hybrid)
2. Cluster (Localization of Anomalies)
3. Market Basket Analysis (New Clustering and Association Rules)
4. Find Laws (Symbolic Knowledge Acquisition Technology ­ SKAT)
5. Memory Based Reasoning (k-Nearest Neighbor and Genetic Algorithms)
6. Find Dependencies (n-Dim joint distribution analysis)
7. Classify (Fuzzy Logic modeling)
8. Discriminate (Unsupervised classification)
9. Stepwise Linear Regression
10. Summary Statistics

These DCOM modules can be readily built into external applications such as different RDBMS’s or Data Warehousing systems, Matlab, or Microsoft Office products with virtually no programming involved.

Utilizing PolyAnalyst COM modules is beneficial to the user who might want to:

* pay only for those machine learning components which are necessary
* keep using a favorite application interface while accessing new
* readily exchange information between different applications
* run applications on a remote computer
* use distributed data analysis applications

Utilizing PolyAnalyst COM modules is beneficial to the integrator or developer who might want to:

* develop new and powerful applications quickly and easily
* incorporate third party components into their applications
* choose the best components from different vendors
* extend the functionality of existing applications by simply adding new components
* combine components written in different environments and languages in a single application
* carry out the development with the most simple and common tools (such as VBA)

According to many analysts, the advent of a well-developed market for software components is going to dramatically influence the future of programming. Megaputer is proud to be among the pioneers in providing a broad selection of DCOM-components for data mining ­ a true interface-oriented programming solution. The new PolyAnalyst COM modules meet a growing interest from well-known integrators and developers of decision support solutions.

“PolyAnalyst analytical engines do an excellent job of finding relations amongst many fields without overfitting,” notes Timothy E. Nagle, Consulting Scientist to 3M. “Megaputer support is outstanding. Inevitable problems one expects with a complex system are dealt with immediately.”

Megaputer offers assistance in integrating our DM components into external applications when desired. A FREE evaluation copy of PolyAnalyst 4.0, the new system built on the described DCOM modules, is available for downloading at

The PolyAnalyst COM toolkit is accompanied by a well documented description of all the provided interfaces and their respective methods.

Platforms: Microsoft Windows

PolyAnalyst is a multi-strategy data mining solution utilizing the latest achievements in automated Knowledge Discovery in Databases. A broad selection of exploration engines allows the user to predict values of continuous variables, explicitly model complex phenomena, determine the most influential independent variables, solve classification and clustering tasks, and process transactional data. The ability to present the discovered relations in explicit symbolic form has no world analogs. The DCOM-based design, point-and-click GUI, versatile data manipulation, visualization, and reporting capabilities, minimum of explicit statistics, and a simple interface to various data storage architectures and external applications make PolyAnalyst a very easy-to-use system.

Megaputer provides a complete family of innovative solutions for Knowledge Discovery in Databases, Data Mining, Semantic Text Analysis, and Information Management. The offered solutions help users make better business decisions by revealing knowledge hidden in their database or textbase. Megaputer customers range from Fortune 100 companies, to numerous smaller businesses, to government offices, and to educational institutions.

Contact: Sergei Ananyan
Megaputer Intelligence
Tel: 812-325-3026; FAX: 812-339-1646

[ Home | About Nautilus | Case Studies | Partners | Contact Nautilus ]
[ Subscribe to Lists | Recommended Books ]

logo Copyright © 1999 Nautilus Systems, Inc. All Rights Reserved.
Mail converted by MHonArc 2.2.0