|
Home
Description
Organizing Committee
Symposium
Dates, times, and places
Why a symposium now?
Who should attend?
Format
List of Accepted Papers
Paper Submission
About AAAI
Decision
Making in M&D
Background
Examples
Artificial Intelligence
Challenge problems
Software and Data
References
FAQ
Web sites
Bibliography
Contact
Comments
Subscriptions
|
Current software
Several commercial and shareware software packages
facilitate classification and segmentation techniques which can be used in
the context of M&D. This (highly incomplete) list includes decision tree
approaches, rule based approaches, neural approaches, and Bayesian Belief Net
approaches.
Decision Tree Approaches
commercial:
- AC2, provides graphical tools for
data preparation and builing decision trees.
- Alice d'Isoft 6.0, a streamlined
version of ISoft's decision-tree-based AC2 data-mining product, is
designed for mainstream business users.
- Business
Miner, data mining product positioned for the mainstream business
user.
- C5.0/See5, constructs classifiers
in the form of decision trees and rulesets. Includes latest innovations
such as boosting.
- CART 4.0 decision-tree software,
from winners in the KDD Cup
2000. Advanced facilities for data mining, data pre-processing and
predictive modeling including bagging and arcing.
- Cognos
Scenario, allows you to quickly identify and rank the factors that
have a significant impact on your key business measures.
- Decisionhouse, provides data
extraction, management, pre-processing and visualization, plus customer
profiling, segmentation and geographical display.
- KnowledgeSEEKER, high performance
interactive decision tree analytical tool.
- Neusciences aXi.DecisionTree,
ActiveX Control for building a decision tree. Handles discrete and
continuous problems and can extract rules from the tree.
- PolyAnalyst, includes an
information Gain decision tree among its 11 algorithms
- SPSS AnswerTree, easy to use
package with CHAID and other decision tree algorithms. Includes decision
tree export in XML format.
- XpertRule Miner (Attar Software),
provides graphical decision trees with the ability to embed as ActiveX
components.
free:
- C4.5, the
"classic" decision-tree tool, developed by J. R. Quinlan,
(restricted distribution)
- EC4.5, a more efficient
version of c4.5, which uses the best among three strategies at each node
construction.
- IND,
provides Gini and C4.5 style decision trees and more. Publicly available
from NASA but with export restrictions.
- LMDT,
builds Linear Machine Decision Trees (based on Brodley and Utgoff
papers).
- ODBCMINE, shareware
data-mining tool that analyzes ODBC databases using the C4.5, and
outputs simple IF..ELSE decision rules in ascii.
- OC1,
decision tree system continuous feature values; builds decision trees
with linear combinations of attributes at each internal node; these
trees then partition the space of examples with both oblique and
axis-parallel hyperplanes.
- PC4.5, a parallel version
of C4.5 built with Persistent Linda (PLinda) system.
- PLUS, Polytomous
Logistic regression trees with Unbiased Splits, (Fortran 90).
commercial:
- Analytica, influence
diagram-based, visual environment for creating and analyzing probabilistic
models (Win/Mac).
- AT-Sigma Data Chopper,
for analysis of databases and finding causal relationships.
- Bayesware Discoverer 1.0,
an automated modeling tool able to extract a Bayesian network from data
by searching for the most probable model
 Data
Digest Business Navigator 5, combines Bayesian networks, graphical UI,
and data preparation tools. - DXpress, Windows based tool for building and
compiling Bayes Networks.
- Ergo(tm), Bayesian Network
Editor and Solver (Win and Mac demos available)
- HUGIN, full suite of Bayesian Network reasoning
tools
- KnowledgeMiner , uses self-organizing
neural networks to discover problem structure (Mac platform)
- Netica, bayesian network tools (Win 95/NT), demo
available.
- PrecisionTree, an add-in for
Microsoft Excel for building decision trees and influence diagrams
directly in the spreadsheet
free:
- BAYDA 1.0
- Bayesian belief network software
(Win95/98/NT/2000), including
- BN PowerConstructor: An efficient system for
learning BN structures and parameters from data. Constantly updated
since 1997.
- BN PowerPredictor: A data mining program for
data modeling/classification/prediction. It extends BN PowerConstructor
to BN based classifier learning.
- FDEP, induces functional dependencies from a
given input relation. (GNU C).
- GeNIe,
decision modeling environment implementing influence diagrams and
Bayesian networks (Windows). Has over 2000 users.
- JavaBayes
- MSBN: Microsoft Belief Network Tools,
tools for creation, assessment and evaluation of Bayesian belief
networks. Free for non-commercial research users.
- Pulcinella, tool for
Propagating Uncertainty through Local Computations based on the Shenoy
and Shafer framework. (Common Lisp)
- SPI, Probabilities, Local Expression Language
Utilities, Explanation, Dynamic Models, GUI. (Common Lisp).
- RoC (Robust Bayesian Classifier) v 1.0, for MS
Windows 9x/NT
Rule-based approach
commercial:
- AIRA, a rule discovery, data and
knowledge visualization tool. AIRA for Excel extracts rules from
MS-Excel spreasheets.
- Datamite, enables rules and knowledge
to be discovered in ODBC-compliant relational databases.
- DataDowser, finds IF[AND]
THEN association rules; uses fuzzy logic.
- PolyAnalyst,
builds fuzzy logic classification rule with PolyNet Predictor, SKAT, or
Linear Regression.
- SuperQuery, business Intelligence tool;
works with Microsoft Access and Excel and many other databases.
- WizWhy, automatically finds all the
if-then rules in the data and uses them to summarize the data, identify
exceptions, and generate predictions for new cases.
- XpertRule Miner (Attar Software)
provides association rule discovery from any ODBC data source.
free:
- CBA, mines association rules
and builds accurate classifiers using a subset of association
rules.
- Claudien,
a clausal discovery engine
- CN2,
inductively learns a set of propositional if...then... rules from a set
of training examples by performing a general-to-specific beam search
through rule-space.
- DBPredictor
- KINOsuite-PR
extracts rules from trained neural networks.
- RIPPER, a
system that learns sets of rules from data. Fast -- asymptotically
O(n*logn*logn), where n is the number of cases. ANSI C, Unix. For
research purposes only.
Neural
Net Approaches:
commercial:
- EMSL
List of Commercial NN tools (58)
- Neural Network FAQ list,
maintained by Warren Sarle of SAS.
- BioComp
iModel(tm), self-optimizing, non-linear predictive model
development.
- COGNOS 4Thought, predictive modeling
tool offering Effectiveness measurement, What-if analysis, and
Forecasting
- BrainMaker, a fast neural network
system, now with MMX acceleration (Win 95/NT)
- DB Prophet
- KINOsuite
- MATLAB Neural Net
Toolbox, a comprehensive environment for neural network research,
design, and simulation within MATLAB
- Neural Innovation Proforma, powerful
predictive model building software suite.
- NeuralWare NeuralSim (formerly
Predict) and Professional II/PLUS.
- NeuroSolutions, featuring
recurrent backpropagation, C++ code generation, and more.
(Win/Unix).
- SPSS Neural
Connection 2, now has Bayesian Network, Data Output Tool, model
weights and more.
- STATISTICA Neural networks,
comprehensive and user-friendly nn application with many charting
options, network architectures and training algorithms.
free and shareware:
Hybrid and other approaches
commercial:
- Affinium Model Suite, includes
linear regression, logistic regression, CHAID, neural networks, and
genetic algorithms.
- Clementine from SPSS, leading
visual rapid modeling environment for data mining. Now includes
Clementine Server.
- Darwin
(now part of Oracle), high-performance data mining software, optimized
for parallel servers
- KINOsuite PR,
extracts rules from trained neural networks.
- Knowledge Studio, featuring multiple
data mining models in a visual, easy-to-use interface.
- MarketMiner automatically selects
the best mining technique using: Statistical Networks, Logistic and
linear regression, K nearest neighbors, and Decision trees (C4.5).
- Polyanalyst, features multiple
classification algorithms: Decision Trees, Fuzzy Logic, and Memory Based
reasoning.
- PredictionWorks,
includes decision tree (gini, entropy, C4.5), logistic regression, k
nearest-neighbor, naive bayes and linear regression. Free test over the
web!
- Previa Classpad, provides an
interactive environment for classification using neural networks,
decision trees, and bayesian networks.
- prudsys DISCOVERER:
non-linear decision trees (NDTs) and sparse grid methods for
classification
- Datalogic, professional tool for
knowledge acquisition, classification, predictive modelling based on
rough sets.
- KXEN, Components, based on Vapnik's work
on SVM.
- K-DYS, rough set approach
- Discipulus, uses genetic
programming for engineering and business data mining problems.
- Evolver, genetic programming
- MARS, J.
Friedman's automated logistic regression for binary classification
problems. Automatic missing value handling, interaction detection,
variable transformation.
- WINROSA, automatically generates fuzzy
rules, based on the fuzzy ROSA method (Rule Oriented Statistical
Analysis).
free:
- BSVM, a decomposition
method for support vector machines (SVM) for large classification
problems.
- LIBSVM, a support
vector machines (SVM) library for classification
- Kernel Machines and related
methods website
- Grobian,
user-friendly software to analyse data with rough set technology (C++).
 Rosetta, is a toolkit
for analyzing tabular data within the framework of rough set theory.
Runs under Windows NT/98/95/2000 - Rough Enough, data mining
tool that supports the entire process; developed under Paradox DBMS for
Windows.
- PEBLS, a nearest-neighbor
learning system designed for symbolic feature values; was applied to
bioinformatic and genetic problems
- TiMBL 2.0, nearest neighbour
approach.
- MLC++, a library of C++ classes
for supervised machine learning, including multiple classification
algorithms.
- JAM, Java Agents
for Meta Learning (applications to fraud and intrusion detection).
- SIPINA-W,
produces decision graphs and trees. Includes several classification
methods. (Win). Shareware
- ROC Convex Hull Program
for comparing classifiers, (perl), under GPL.
Useful datasets
Some of the standard datasets which can be used for
benchmarking (such as the IRIS dataset) are available at the machine learning
data set repository at UCI which is accessible via ftp://ftp.ics.uci.edu/pub/machine-learning-databases/
|
