Togaware: Data Mining Catalogue

Data Mining Catalogue

Welcome to the Catalogue of Data Mining tools and service providers. The material in this catalogue is migrating to form a new book, The Data Mining Desktop Survival Guide.

Many Data Mining tools, vendors, and service providers are now available to service the Data Mining market. This Catalogue provides pointers to Data Mining tool vendors and service providers. The information is provided as is, without any guarantees. If information is incorrect or out of date, or you have information relating to other Data Mining tools or services, please let me know (Graham.Williams@togaware.com).

This compilation is Copyright © 2000-2006 Graham J. Williams. Redistribution of this catalogue is permitted under the general terms and intensions of the GNU General Public License. This catalogue is provided for reference only. Neither I nor Togaware, endorse nor in any way recommend the products and organisations listed and expressly exclude liability for any damage, loss or injury that a person may suffer as a result of any dealing with any product or organisation listed.

Free Data Mining Software

Bayesian Knowledge Discoverer

Techniques: Bayesian Networks

Platforms: Win32

Vendor: Freely available from the Open University

Internet: kmi.open.ac.uk/projects/bkd/

Updated 1999/11/30 13:39:42 Graham.Williams@togaware.com

PLUS

Polytomous Logistic regression trees with Unbiased Split. PLUS is implemented in a set of Fortran 90 routines and accepts numerical/continuous as well as categorical variables. Missing covariate values are allowed. If a test data set is available, an estimate of the misclassification error rate will be provided. Executables are available for Digital Alpha (Digital UNIX 4.0), Sun SPARCstation/Ultra (Sun Solaris 2.6), Pentium (Linux), and Pentium (Windows 95/98/NT, coming soon).

Techniques: Logistic Regression

Platforms: Unix, Linux, Win32

Author: Tjen-Sien Lim

Internet: http://www.recursive-partitioning.com/plus/software.html

Updated 1999/11/01 09:36:41 Graham.Williams@togaware.com

Rattle

An open source graphical data mining tool written in R and ideally suited for binary classification tasks and quick data mining.

Techniques: decision trees, random forests, support vector machines, boosting, logistic regression, kmeans

Platforms: Unix, Linux, Win32, MacOS

Author: Graham Williams

Internet: http://rattle.togaware.com

Updated 2006-07-31 07:00:36 Graham Williams

Weka
See The Data Mining Desktop Survival Guide

Vendors and Service Providers

4cData

4cData offers two state of the art DM tools:

4cRuleBuilder;
4cDiscretizer.

The products are characterized by easy to use GUI, and availability of free trial versions.

4cData also provides a full range of data mining and knowledge discovery services including:

building a data model
generating classification rules to perform a variety of tasks
discovering major trends in your data
performing data prediction
providing statistical analysis of the data

Techniques:	decision rules, discretization
Platforms:	Win95, Win98, WinNT, Win2000, WinXP
Vendor:	4cData, Golden, CO
Internet:	www.4cdata.net

Updated 2004/01/23 09:06:16 Graham.Williams@togaware.com

Angoss

Angoss Software Corporation (Angoss), headquartered in Toronto, Canada, is a provider of Predictive Analytics Systems, for sales, marketing and risk management.

Angoss offer:

KnowledgeSTUDIO: multiple data mining models
KnowledgeSEEKER: high-performance decision tree tool
KnowledgeSERVER: deploy models, scoring, and rules
StrategyBUILDER: manage models and user business rules

Techniques:	decision trees, k-means clustering, neural networks, linear and logistic regression
Platforms:	Win32, Unix
Vendor:	Angoss
Internet:	www.angoss.com

Updated 1999/12/16 06:09:20 Graham.Williams@togaware.com

Bayesia: BayesiaLab

Bayesia is a company created by researchers specialising in Bayesian Networks and Machine Learning. The commercial product, BayesiaLab, offers a set of functionalities for modeling and data mining:

Ergonomic graphical interface allowing creating your models by simple clicks, with equation editor for a compact description of your probability distributions
Complete set of Bayesian networks learning algorithms: Supervised learning algorithms for profiling target variables, unsupervised learning for discovering all the probabilistic relations that hold your data, Clustering for new concept discovery
Data importation Wizards allowing data preprocessing
Missing values and hidden variables processing
Evaluation of the models with confusion matrices, lift and ROC curve
Analysis toolbox for easy models' understanding: strength and type of the probabilistic relations, sensitivity analysis, causal analysis, contradiction analysis, HTML reports generation
Powerful automatic layout algorithms for both tree networks and highly connected and complex graphs
High interoperability, with JDBC/ODBC connections to data bases, SQL interface, and exportation of Bayesian networks, tables, equations, graphs, matrices and reports just by copying and pasting.
Dynamic Bayesian networks for taking into account the temporal dimension
Decision Aiding tools with Decision and Utility nodes, direct representation of action policies, and reinforcement learning algorithms for automatic discovering of policies for static as well as dynamic Bayesian networks

Techniques:	Bayesian nets
Platforms:	Win32
Vendor:	Bayesia
Internet:	www.bayesia.com

Updated 2004/02/17 08:54:07 Graham.Williams@togaware.com

Equbits Foresight
See The Data Mining Desktop Survival Guide

Fujitsu: GhostMiner
See The Data Mining Desktop Survival Guide

OpenMIND

A company whose primary business is consulting in commercial data mining & VLDB decision support across Europe. Provide services in Technical Strategy and Tool Selection, Data Mining projects using CRISP-DM, Data Warehousing management consulting, technical design and implementation. Provide bureau services in Data Mining and Visualisation, and Web Information Publishing.

Markets: Communication, Finance, Retail, Manufacturing.

Training: Running Data Mining Projects using CRISP-DM
Understanding Data Mining in Business

Vendor: OpenMIND Consulting, Oxfordshire, UK

Internet: www.open-mind.co.uk/impg/impg.htm

Updated 1999/12/15 15:20:51 Graham.Williams@togaware.com

PredictionWorks: InductionEngine
See The Data Mining Desktop Survival Guide

Purple Insight: MineSet

Purple Insight acquired the MineSet product from SGI in October, 2003. MineSet provide interactive exploration of data through an advanced suite of visual tools for faster discovery of meaningful trends and relationships. The Splat Visualizer and the Scatter Visualizer represent complex data in up to eight dimensions. The Map Visualizer displays data with geographical relationships by using a map metaphor. Animation and view synchronization techniques are used to reveal patterns over critical dimensions such as time. The Tree Visualizer depicts data with hierarchical relationships utilizing a fly-through technique set in a 3D landscape. The Statistics Visualizer presents a visual summary of basic statistical information. Advanced drill-through techniques give you fast access to the original records that created entities within your visualization for additional analysis. The Record Viewer allows quick access to the original data with column sorting and HTML output capabilities.

Techniques: mlc++, visualisation, rules, associations, clustering

Platforms: MSWindows, IRIX

Vendor: Purple Insight

Internet: www.mineset.com

Updated 2004/02/18 06:03:14 Graham.Williams@togaware.com

RapAnalyst

RapAnalyst allows you to visualize and mine complex data.

Techniques: neural networks, genetic algorithms, self organizing maps

Platforms: MSWindows

Vendor: Raptor International

Internet: raptorinternational.com

Updated 2006-01-06 15:30:27 Graham.Williams@togaware.com

Sherrod

Phil Sherrod has developed two statistical programs:

DTREG generates classification and regression decision trees. It uses V-fold cross-valication with pruning to generate the optimal size tree, and it uses surrogate splitters to handle missing data. A free demonstration copy is available;
NLREG performs general nonlinear regression. NLREG will fit a general function, whose form you specify, to a set of data values. A free demonstration copy is available.

Techniques:	classification and regression decision trees, nonlinear regression
Platforms:	MSWindows
Author:	Phil Sherrod
Internet:	www.dtreg.com for DTREG and www.nlreg.com for NLREG

Updated 2004/01/23 09:06:16 Graham.Williams@togaware.com

from Daz Systems

An agent-oriented knowledge discovery system.

Vendor: Daz Systems, Inc, California

World Wide Web: www.dazsi.com

Updated 24 Jul 1997 09:41:18 Graham.Williams@togaware.com

C5.0 and See5

This is an updated, commercial, and significantly faster, version of C4.5, a very popular decision tree induction algorithm.

Techniques: Decision Trees, Rules

Platforms: Unix, Win95, WinNT

Vendor: RuleQuest, Sydney

World Wide Web: www.rulequest.com

Updated 25 Jul 1997 09:39:24 Graham.Williams@togaware.com

CART^®

CART^® (Classification and Regression Trees) is a widely used and respected Decision Tree induction and Regression system from the Statistical community with many applications to Data Mining.

Techniques: Classification and Regression Trees with Boosting

Applications: classification and predictions

Platforms: Win95, MacOS, Unix

Vendor: Salford Systems, San Diego, California

World Wide Web: www.salford-systems.com

Updated 9 Sep 1998 10:01:36 Graham.Williams@togaware.com

from SPSS

See The Data Mining Desktop Survival Guide

Cube-It

Techniques: OLAP, Visualisation

Market: Finance

Vendor: Financial Information Consulting Services, Belgium

World Wide Web: www.ficsgrp.com

Updated 30 Jul 1997 07:38:27 Graham.Williams@togaware.com

from Thinking Machines

See Oracle Data Mining in The Data Mining Desktop Survival Guide

DataCruncher

DataMind DataCruncher is a server based data mining software product. Able to handle an unlimited number of records, DataCruncher supports advanced data mining features such as incremental modeling, model merge, and native connectivity to Oracle and Informix database management systems.

Techniques: Agent Network Technology

Market: Telecomunications Marketing (Telecom Customer Life Cycle)

Platforms: Server: Sun, HP, IBM, SGI, and WinNT. Client: WinNT and Win95

Vendor: DataMind Corporation, San Mateo, California

World Wide Web: www.datamindcorp.com

Updated 10 Oct 1997 20:06:47 Graham.Williams@togaware.com

DataEngine combines conventional data analysis methods with fuzzy logic and neural networks.

Techniques: Fuzzy Logic, Neural Networks, Fuzzy Clustering.

Markets: Industry, Trade, Financial Services.

Applications: Quality Control, Process Optimization Forecasting, Database Marketing, Diagnosis

Platforms: Win95, WinNT

Vendor: MIT - Management Intelligenter Technologien GmbH, Aachen, Germany

World Wide Web: www.mitgmbh.de

Updated 30 Jan 1998 12:33:54 Graham.Williams@togaware.com

from Logic Programming Associates

A tool for investigating data from ODBC-compliant relational databases through the synthesis of clusters and rule induction with queries direct to the database engine.

Techniques: Rule Induction, Clustering

Platforms: WinNT, Win95

Vendor: Logic Programming Associates (LPA), England

World Wide Web: www.lpa.co.uk

Updated 22 Jul 1998 06:27:56 Graham.Williams@togaware.com

The Data Mining Suite works directly on large SQL repositories with no need for sampling nor to extract files. It accesses large volumes of multi-table relational data on the server, incrementally discovers patterns and delivers automatically generated English text and graphs as explainable documents on the intranet.

Techniques: Rule-based Influence; Dimensional Affinity; Trends

Vendor: Information Discovery, Inc, Hermosa Beach, California

World Wide Web: www.datamining.com

Updated 13 Mar 1998 14:38:47 Graham.Williams@togaware.com

DBMiner

An OLAM tool supporting Data Cubes. An educational version (restricted to 1000 rows of data) available for free.

Techniques: OLAP, Associations, Decision Trees, Regression

Platforms: Win32

Vendor: DBMiner Technology Inc, Canada

World Wide Web: www.dbminer.com

Updated 18 Jul 1998 17:16:08 Graham.Williams@togaware.com

dbProbe is a lightweight Java Applet that enables multi-dimensional cube analysis via any of the leading web browsers. The user is presented with an effective, true GUI rather than an HTML-based dialoge. The server-based cube-builder component can access various data sources including ODBC and OLE/DB. dbProbe uses a unique architecture which enables the whole cube to be transferred to the client allowing off-line analysis and a highly scalable server model. A client-server model is also available to allow analysis of massive cubes.

Techniques: Multi-dimensional cube construction and analysis, GUI representation, Web-based deployment

Platforms: MS-WindowsNT, MS-Windows95, Linux, UNIX

Vendor: InterNetivity Inc, Canada

World Wide Web: http://www.internetivity.com

Distributor:

Updated 1999/06/11 15:59:01 Graham.Williams@togaware.com

dbProphet

Techniques: Neural Network

Market: Sales and Marketing

Applications: Business retention, resource allocation, prospecting, cross selling and upselling, market basket analysis, pricing, and fraud detection

Platforms: WinNT, Unix

Vendor: Trajecta, Austin, Texas

World Wide Web: www.trajecta.com

Updated 28 Jan 1998 09:08:53 Graham.Williams@togaware.com

Decisionhouse is a scalable data mining product for (marketing and risk) customer modelling applications in finance, retial and telco markets.

Techniques: Visualisation, Selection, Drill-Down, Decision Trees, Discriminant Analysis, Logistic and Probit Regression, Gini and Kolmogorov-Smirnov Analysis

Markets: Finance, Retial, Telecommunications

Applications: Customer profiling, targeting, segmentation, assessment of risk, customer worth, lifetime value, purchasing propensity and churn modelling.

Users: Barclays Bank, J Sainsbury plc, Liverpool Victoria Group, Barclaycard, British Airways

Platforms: Unix Servers; Unix and Win95 Clients

Vendor: Quad Stone, Scotland

World Wide Web: www.quadstone.com

Updated 19 Feb 1998 12:02:16 Graham.Williams@togaware.com

Discovery Server

Techniques: Segmentation

Market: Sales, Marketing

Platforms: Win95, WinNT

Vendor: Pilot Software, Massachusetts

World Wide Web: www.pilotsw.com

Updated 25 Jul 1997 10:12:25 Graham.Williams@togaware.com

from

See The Data Mining Desktop Survival Guide

Built on top of SAS, GainSmarts provides a statistically-based approach for the exploration of patterns and data.

Techniques: CHAID, Genetic Algorithms, Predictive Modelling, Expert Systems

Markets: Direct Marketing

Platforms: SAS, WinNT, Unix

Vendor: Urban Science, Detroit, Michigan

World Wide Web: www.urbanscience.com

Updated 3 Apr 1998 16:25:51 Graham.Williams@togaware.com

Intelligent Miner

A collection of tools to handle organisational/transactional data for Data Mining.

Techniques: Association, Segmentation, Time Sequence, NN, Rules

Platforms: AIX, AS/400, OS/390, MVS, Win95, WinNT, OS/2

Vendor: IBM

World Wide Web: www.software.ibm.com/data/intelli-mine/

Updated 24 Jul 1997 09:58:21 Graham.Williams@togaware.com

from AcknoSoft

Techniques: Decision Trees, Case-Based Reasoning

Vendor: AcknoSoft, Paris

World Wide Web: www.acknosoft.com

Updated 20 Feb 1998 17:01:17 Graham.Williams@togaware.com

Market: Retail

Vendor: Knowledge Discovery One, Inc, Texas

World Wide Web: www.kd1.com

Updated 30 Jul 1997 16:02:57 Graham.Williams@togaware.com

MARS^® (Multivariate Adaptive Regression Splines) was developed by Jerry Friedman and is another of his widely respected data mining tools with sound statistical foundations.

Techniques: Multivariate Adaptive Regression Splines

Applications: classification and prediction

Vendor: Salford Systems, San Diego, California

World Wide Web: www.salford-systems.com

Updated 19 Sep 1998 09:28:43 Graham.Williams@togaware.com

Model 1

A Windows-based family of data mining modules for database marketing applications. Model 1 includes a wizard-driven GUI, automated model building process using a variety of algorithms, easy-to-interpret reports, built-in campaign optimization, and run-time module deployment.

Applications: Response Modeling, Cross-Selling, Customer Valuation, Segmentation, Profiling

Platforms: Win32

Vendor: Unica

World Wide Web: www.unica.com

Updated 19 Sep 1998 06:29:25 Graham.Williams@togaware.com

ModelQuest

Products include ModelQuest MarketMiner and ModelQuest Enterprise. Techniques automatically capture relationships in data using a network of mathematical functions that subdivide problems into manageable pieces, or nodes, and automatically apply regression techniques to solve each of these simpler problems. ABTech also offer Data Mining services.

Techniques: Neural Nets, Regression

Platforms: Win95, WinNT

Vendor: ABTech

World Wide Web: www.abtech.com

Updated 14 Nov 1997 18:08:18 Graham.Williams@togaware.com

Decision Series

Techniques: Neural Nets, Association Rules, Clustering, Genetic Algorithms

Market: Retail and market basket analysis

Platforms: SMP

Vendor: Neovista

World Wide Web: www.neovista.com

Updated 28 Aug 1997 13:57:10 Graham.Williams@togaware.com

Offer a variety of products and services for data mining.

Techniques: NN

Market: Telecom Fraud, Market Churn, Credit Scoring

Vendor: Neural Technologies, England

World Wide Web: www.neuralt.com

Distributor: Neural Mining Solutions Pty Ltd (Australia)

Updated 25 Jul 1997 11:35:22 Graham.Williams@togaware.com

ODBCMine

ODBCMine generates decision rules from ODBC databases using the C4.5 classification model algorithm. It analyzes the data in any ODBC data source, and writes decision rules in ASCII to the standard output device. The 2.0 release outputs graphical decision trees in Scalable Vector Graphics (SVG) format. It also provides a new method for users to pre-categorize continuous (numeric) variables into discrete ranges, providing much better performance for these situations. It is intended to be a simple and inexpensive - yet powerful - implementation of a classic data mining algorithm. A single user-license is USD $99.95. An academic license is available for free by request.

Techniques: Decision Tree Induction

Platforms: MSWindows

Vendor: Intelligent Systems Research, Chicago

World Wide Web: www.xmlmethods.com

Updated 2003/03/31 Graham.Williams@togaware.com

Orchestrate

A scalable, parallel, large volume data management tool, providing a data mining platform.

Techniques: Data Management, Statistics, Neural Networks, Decision Tree Induction

Platforms: Unix, Unix SMP

Vendor: Torrent Systems, Inc, Massachusetts

World Wide Web: www.torrent.com

Updated 4 Mar 1998 13:16:30 Graham.Williams@togaware.com

Partek

A comprehensive pattern recognition and data visualization system. Direct manipulation visual data mining combined with statistical, neural, and other pattern recognition techniques. Capabilities: Interactive Visualization, Statistical Analysis, Normalization & Transformation, Variable Selection, Model Development & Deployment.

Platforms: Unix (SGI, Sun, HP, Linux), Windows 9x/NT

Vendor: Partek Inc, Missouri

World Wide Web: www.partek.com

Updated 24 Jul 1997 08:37:51 Graham.Williams@togaware.com

PATTERN

Magnify is a provider of data mining services with their own sophisticated tool for data mining very large datasets.

Platforms: Unix

Vendor: Magnify Inc, Chicago

World Wide Web: www.magnify.com

Updated 24 Jul 1997 08:37:51 Graham.Williams@togaware.com

PolyAnalyst is a multi-strategy data mining environment with a selection of exploration engines to predict values of continuous variables, explicitly model complex phenomena, determine the most influential independent variables, and solve classification and clustering tasks. Features include object-oriented design, point-and-click GUI, versatile data manipulation, visualization, and reporting capabilities, minimum of statistics, and a simple interface to various data storage architectures.

Products: PolyAnalyst 4.0 (MSWindows)

Techniques: Genetic Algorithms, Multidimensional Analysis, Stepwise Linear Regression, Statistics, Localization of anomalies, Neural Networks, Associations, Memory-Based Reasoning

Platforms: Win32

Market: Finance, Marketing, Medicine, Telecom, Retailing

Vendor: Megaputer Intelligence, Indianna

Distributor: Megaputer

World Wide Web: www.megaputer.com

Updated 2000/01/21 08:30:39 Graham.Williams@togaware.com

Q-Why

A freeware (academic/personal use) rule finding tool for small to medium data mining tasks, with a commercial tool available for larger tasks.

Techniques: Rules

Platforms: Win95, WinNT

Vendor: Quintillion Corporation

World Wide Web: www.qwhy.com/qwhy/

Updated 24 Jul 1997 10:24:43 Graham.Williams@togaware.com

Scenario

Techniques: Statistics, Segmentation, Rules, Visualisation

Platforms: Win95

Vendor: Cognos, Canada

World Wide Web: www.cognos.com/busintell/products/scenario.html

Updated 24 Jul 1997 12:07:37 Graham.Williams@togaware.com

Data Mining

"SPSS offers an integrated family of products and services that makes it easy to share intelligence found via data mining. From simple to sophisticated analysis, SPSS offers a range of products for modeling."

Techniques: CHAID, decision trees, neural networks, data visualization, multidimensional reporting, extensive range of statistics, plus statistical components.

Platforms: Win95, WinNT, UNIX

Vendor: SPSS, Chicago, IL

World Wide Web: www.spss.com/datamine/

Press: DM Review

Updated 6 Aug 1998 06:41:10 Graham.Williams@togaware.com

STATISTICA Data Miner

See The Data Mining Desktop Survival Guide

SuperQuery

Techniques: Rule Induction

Platforms: Win95, WinNT

Vendor: Azmy Thinkware Inc.

World Wide Web: www.azmy.com

Updated 24 Jul 1997 07:46:17 Graham.Williams@togaware.com

Object Relational Data Mining

Tandem provides the platform for running a variety of partner data mining products.

Vendor: Tandem, Silicon Valley

World Wide Web: www.tandem.com

Updated 25 Jul 1997 11:04:47 Graham.Williams@togaware.com

Virtual Predict

See The Data Mining Desktop Survival Guide

Visualizer Workstation

Techniques: N-dimensional graphics, feature extraction and enhancement, sample, autocluster, descriptive knowledge-based trainer

Platforms: WinNT

Vendor: Computer Science Innovations, Inc, Melbourne, Florida

World Wide Web: www.csihq.com

Updated 29 Jan 1998 15:33:11 Graham.Williams@togaware.com

Techniques: Rules

Platforms: Win95, WinNT

Vendor: WizSoft

World Wide Web: www.wizsoft.com

Updated 24 Jul 1997 09:15:37 Graham.Williams@togaware.com

XpertRule

XELOPES is a an open platform-independent and data-source- independent library for Embedded Data Mining. It is based on the CWM standard of the OMG and supports other Data Mining standards like PMML, OLE DB for DM, and MLC++. It is available under Java, C++, and C#.

Techniques: Decision Trees, Parallel SQL

Platforms: Win95, WinNT

Vendor: Attar Software, England

World Wide Web: www.attar.com