[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Demonstrations of Advanced Statistical Analysis Tools, CART, MARS,MART, Random Forests, onsite meetings May 23 - May 30
The CEO of Salford Systems, Dr. Dan Steinberg, will be visiting
Italy, May 23-30. He will be demonstrating our latest data mining tools (based on the CART, MARS, MART and Random Forests algorithms). We have been working with the original developers of CART and MARS software since 1990.
Below is a list of the presentation titles/speakers from our recent
user group conference. This will give you an idea of how our CART, MARS, MART and
Random Forests tools are currently used for modeling.
Also below is some information about CART, MARS, TreeNet and Random Forests.
Best Regards,
Lisa Solomon
lisas@salford-systems.com
619-543-8880 x14
http://www.salford-systems.com?italy
PRESENTATION TITLES FROM CART DATA MINING 2004
David Goldsmith, MDT Advisers,
"Combined Time Series and Cross Sectional CART Modeling for Common Stock
Selection"
Cecil Dharmasri, Syngenta Crop.,
"The Importance of CART and MARS in Environmental Fate and Risk Assessment
for Pesticides"
Joseph Cappelleri, Pfizer,
"Using CART to Develop a Diagnostic Tool for Erectile Dysfunction"
Charles Pollack, Suncorp Metway,
"Insurance Premium Increase Optimization: Case Study"
Belle Hudischewskyj, ICF Consulting/Systems Applications International,
Inc.
"Application of CART for Air Quality Forecasting"
Marsha Wilcox, Boston University Medical School
"Using CART to Discern Models in Genetics: Alcoholism, Alzheimers and
Aging"
John Trimble, Wells Fargo
"CART/ MARS Risk Assessment in Automobile Loans and Leases"
Falk Heuttmann, University of Alaska
"Predictive and Spatial Modeling Applications for Wildlife Research and
Conservation Management: An Overview Linking CART and MARS with GIS"
Jason Haukoos MD, Denver Health Medical Center
"CART for Outcome Predictions in Clinical Settings: Emergency Department
Triage, Survival Prediction and Prediction of Neurologic Survival"
David McCloskey, Pathfinder Solutions Ptd.
"Predicting Customer Behavior Trends Over Space and Time"
Christopher Martius, University of Bonn
"Assessing the Sustainability of Agroforestry Systems using CART to
Model Non-linear Relationships"
Debopriya Das, Cold Spring Harbor Laboratory
"Application of MARS to Gene Expression Data: Predictive Models of Gene
Regulation"
Larry Lai, DIRECTV, Inc.
"Variable Derivation and Selection For Customer Churn Models"
Wayne Danter, Critical Outcome Technologies, Inc.
"Drug Discovery using CART and MARS"
Stuart Gansky, UCSF School of Dentistry
"MARS and Related Techniques in Dental Biomaterials Modeling"
Edward Malthouse, Northwestern University
"High-Level Marketing Strategy: Deciding Whether to Customize Product
Offerings and their Promotion and Whether to Reward “Best” Customers
with Perks"
Donovan Chin, Biogen
"Improving the Ability to Predict Drug-Like Compounds from Virtual
Screening Using CART"
Kenna Mawk, Ciphergen Biosystems, Inc.
"Mining SELDI ProteinChip Data for Biomarkers and Disease Stratification"
Louise Francis, Actuarial Analytics
"Insurance Fraud Detection: MARS vs. Neural Networks"
Jing Wang, Cengent Therapeutics
"CART in Drug Discovery: Identifying Rules for Making Better Small
Molecules"
Brydon Grant, University of Buffalo
School of Medicine
"Using MARS for the Prediction of the Apnea-Hypopnea Index"
John Farar, Union Bank
“Union Bank’s Use of CART to Identify Customer Attrition and Screen
Application Fraud”
John Warner, Novartis
"Drug Discovery Clinical Trials and Random Forests at Novartis"
Shenghan Lai, Johns Hopkins Bloomberg School of Public Health "Examples
in Epidemiology using
CART, MARS and TreeNet"
PRODUCT INFORMATION
Salford Systems' CART is the only classification and regression
tree software based on the original methods and procedures developed
by Breiman, Friedman, Olshen and Stone. We have been working with
these researchers since 1990 to perfect the engine to give you a
celebrated and award-winning system.
Jerome Friedman's MARS (Multivariate Adaptive Regression Splines) is stepwise regression done right for the first time. MARS does variable selection, variable transformation, interaction detection, and self-testing to prevent overfitting, all automatically. Like CART, there is only one trademarked MARS and it is available exclusively from Salford Systems.
TreeNet, Jerome Friedman's latest data mining tool, is based on boosted decision trees. TreeNet is an astonishingly accurate model builder and function approximation system that also serves as a powerful initial data exploration tool. Use TreeNet to extract the most important relationships in your data and calibrate how predictable the outcomes are. Then either use the TreeNet model directly or incorporate the results in CART, MARS, or conventional statistical models.
Random Forests, Leo Breiman's latest data mining technology, is based on learning ensembles of CART trees. By judiciously injecting randomness into the tree building process and then combining hundreds of these trees, RF is able to deliver high performance predictive models and a variety of novel exploratory data analysis results. RF also incorporates new metric free CLUSTER analyses that automatically select the variables used to define each cluster, with potentially different variables defining each cluster.
=============================================================================
NOTA:
Le norme per utilizzare il forum SIS e le istruzioni per iscrizione
e cancellazione sono disponibili all'indirizzo
.
. http://w3.uniroma1.it/sis/forum.asp
.
L'archivio di tutti i messaggi (aggiornato al giorno precedente)
e' disponibile all'indirizzo
.
. http://www.stat.unipg.it/cgi-bin/wilma/sis/
.
=============================================================================