CART^® is a robust, easy-to-use decision tree tool that automatically sifts large, complex databases, searching for and isolating significant patterns and relationships. This discovered knowledge is then used to generate reliable, easy-to-grasp predictive models for applications such as profiling customers, targeting direct mailings, detecting telecommunications and credit card fraud, and managing credit risk.

In addition, CART is an excellent pre-processing complement to other data analysis techniques.  For example, CART's outputs (predicted values) can be used as inputs to improve the predictive accuracy of neural nets and logistic regression.

Product Overview

Features

What's new in CART 4.0 Pro

TreeCoder (Model Deployment Module)

TreeViewer (NEW add-on for UNIX platforms)

Applications and Bibliography

What People are Saying About CART

Customer Success Stories

Frequently Asked Questions

University Program Agreement

CART Training Seminars

CART Monograph

Japanese Textbook on CART

Technical Overview

White Papers

Operating Systems

Scalability

Files Supported

FAQs (Technical Support)

CART WalkAbout -- Quick Tour of CART's Easy-to-Use Interface and Unique Functionality

Features

CART uses an intuitive, Windows based interface, making it accessible to both technical and non technical users.  Underlying the "easy" interface, however, is a mature theoretical foundation that distinguishes CART from other methodologies and other decision trees.  Salford Systems' CART is the only decision tree system based on the original CART code developed by world renowned Stanford University and University of California at Berkeley statisticians; this code now includes enhancements that were co-developed by Salford Systems and CART's originators.

Based on a decade of machine learning and statistical research, CART provides stable performance and reliable results.
Its proven methodology is characterized by:

a reliable pruning strategy,

CART's developers determined definitively that no stopping rule could be relied on to discover the optimal tree, so they introduced the notion of over-growing trees and then pruning back; this idea, fundamental to CART, ensures that important structure is not overlooked by stopping too soon.  Other decision tree techniques use problematic stopping rules.

CART's binary decision trees are more sparing with data and detect more structure before too little data is left for learning.  Other decision tree approaches use multi-way splits that fragment the data rapidly, making it difficult to detect rules that require broad ranges of data to discover.

In the search for patterns in databases it is essential to avoid the trap of "overfitting," or finding patterns that apply only to the training data.  CART's embedded test disciplines ensure that the patterns found will hold up when applied to new data.  Further, the testing and selection of the optimal tree are an integral part of the CART algorithm.  Testing in other decision tree techniques is conducted after-the-fact and tree selection is left up to the user.

surrogate splitters intelligently handle missing values,

CART handles missing values in the database by substituting "surrogate splitters," which are back-up rules that closely mimic the action of primary splitting rules.  The surrogate splitter contains information that is typically similar to what would be found in the primary splitter.  Other products' approaches treat all records with missing values as if the records all had the same unknown value; with that approach all such "missings" are assigned to the same bin.  In CART, each record is processed using data specific to that record; this allows records with different data patterns to be handled differently, which results in a better characterization of the data.

CART can accommodate situations in which some misclassifications, or cases that have been incorrectly classified, are more serious than others.  CART users can specify a higher penalty for misclassifying certain data, and the software will steer the tree away from that type of error.  Further, when CART cannot guarantee a correct classification, it will try to ensure that the error it does make is less costly.  If credit risk is classified as low, moderate, or high, for example, it would be much more costly to classify a high risk person as low risk than as moderate risk.  Traditional data mining tools cannot distinguish between these errors.

CART includes seven single variable splitting criteria - Gini, symmetric Gini, twoing, ordered twoing and class probability for classification trees, and least squares and least absolute deviation for regression trees - and one multi-variable splitting criteria, the linear combinations method.  The default Gini method typically performs best, but, given specific circumstances, other methods can generate more accurate models.  CART's unique "twoing" procedure, for example, is tuned for classification problems with many classes, such as modeling which of 170 products would be chosen by a given consumer.  To deal more effectively with select data patterns, CART also offers splits on linear combinations of continuous predictor variables.

Applications and CART Bibliography

Industries using CART include telecommunications, transportation, banking, financial services, insurance, health care, manufacturing, retail and catalog sales, and education.  Applications span:

market segmentation, customer profiling, retention/attrition analyses

market segment profitability, campaign targeting, response prediction
(e.g., see Cabela's success story)

credit card scoring, fraud detection (e.g., see Fleet Bank success story)

assembly line failures, quality control

clinical trials, biomedical research  (e.g., see Pfizer success story)

Click here to review an extensive bibliography containing over 300 citations.

RETAIL

"CART is an important statistical analysis tool that we use to segment our databases and predict risk factors for the Sears Card.  The advantage of the decision tree format is that our results are easy to interpret; especially with CART, we are able to see a great deal of detail about each of the nodes, such as the node's misclassification costs, the count of data assigned to that node, and a display of the surrogate values substituted for the node."

Steven Li, Senior Manager, Risk Technology, Sears, Roebuck and Co

TELECOMMUNICATIONS

"When we purchased CART, it was the only comprehensive classification and segmentation software available that could handle the large data sets we use for credit card risk management.  In addition, CART provides us with a great deal of flexibility by allowing us, for example, to specify a higher penalty for misclassifying a certain data value."

Feng Xu, Senior Manager, AT&T Universal Card Services

BANKING/FINANCE

"CART offers two distinctive advantages that other database segmentation tools do not.  First, it allows the analyst to identify the smallest target segment possible, such as 10 out of tens of thousands, with exceptional precision.  In addition, CART allows us to specify a higher penalty for misclassifying a potentially poor prospect than for rejecting a good one; this makes us more confident that, for products with very thin margins, our segmentation models avoid prospects who would likely be non profitable.  CART is an invaluable data mining and modeling tool for Fleet Financial Group."

Terence Mak, VP, Lead Analytic Consultant, Fleet Financial Group

GOVERNMENT

"I use CART to provide Canadian meteorologists with dynamic statistical models for predicting lake effect snowfall, ozone levels and other weather issues that affect Canada.  The optimal tree models I create in CART have proven their accuracy many times over when the tree is used with independent data."

William Burrows, Meteorological Research Scientist, Atmospheric Environment Service

 "As a statistician in the Naval environment, I have been involved in the field of data mining for the past four years.  Classification trees has become one of the primary tools with which I extract useful information from large data
bases. I have used various different classification tree software, and have found CART to be the superior product. What I find particularly useful are the
following: 

* The colour codes of the nodes which one can use to pick the most important branches (or rules). 

* The relative cost vs number of nodes graph which I always use to select the "least complicated" with "low" relative cost. 

* The Gains chart provides a good graphical view for assessing tree performance. 

Dr Martin Kidd
IMT
South Africa

MARKETING

"PreVision Marketing's clients include Fortune 500 companies from telecommunications, automotive, retail and packaged goods industries.  We apply our database marketing and analysis expertise to turn our clients' usual wealth of customer information into beneficial marketing information and customer relationship programs.  At PreVision, this typically includes developing models of customer and prospective customer behavior.  CART's recursive partitioning abilities give us a proven statistical method for generating marketing models in an easy-to-understand decision tree format.  This format is accessible to all of our clients, even those with limited statistical backgrounds, and the clarity of the decision tree display gives our clients added confidence in the validity and utility of the models we create."

Marsha Wilcox, Ed.D., Vice President, PreVision Marketing

INDUSTRIAL

"At Chevron, we conduct a lot of exploratory work for oil well drilling.  Instead of taking many expensive core samples, we can use monitoring tools to characterize geographic areas; this data capture generates small data sets with variables that are complex and interrelated rather than independent.  CART, with its v-fold cross validation capability, is our tool of choice for analyzing these small, complex data sets."

Wesley Johnston, Chevron Information Technology Co.

ACADEMIC

"As a research scientist in both academic and professional environments, I work with databases too large and complex to process manually.  CART, unlike multiple linear programming and other methods that are constrained by functional forms, shows me truer characterizations of interrelationships between the data.  CART is also a robust program that can support a diverse set of applications ranging, in my case, from food security analyses to pattern recognition and remote sensing problems."

Eric Weiss, Ph.D., Consultant; Arid Lands Resource Sciences, University of Arizona

Customer Success Stories

Critical Outcome Technologies Uses CART^® and MARS^® to predict which molecular structures are effective against HIV (PDF file)
(From August 2000 issue of PCAI Magazine)

Fleet Uses CART^® Data Mining Technology to Understand Customer Characteristics and Habits

Cabela's® Stabilizes Catalog Mail Model Segments With CART^® Data Mining Software

Pfizer Enlists CART^® to Score Male Erectile Dysfunction Diagnostic Test

University Program Agreement

Salford Systems' University Program provides CART at significantly reduced licensing fees to the educational community.   Eligible educational institutions are colleges, universities, community colleges, technical schools, and science centers.  The University Program gives eligible educational institutions the right to distribute CART right-to-use licenses to all faculty, staff, and students for personal computers, and to install UNIX versions of CART on University workstations and servers.  For more information on this special program, please contact our sales department.

CART Monograph

The definitive  monograph on the theoretical underpinnings of CART by none other than CART's originators, Breiman, Friedman, Olshen and Stone.  Their landmark work created the modern field of sophisticated, mathematically- and theoretically founded decision trees. A must read for those interested in learning about the statistical and technical details.  Three mathematical chapters provide proofs showing what happens in increasingly large data sets; the answer is: CART trees converge to the true structure of the data or yield the true conditional probabilities of classification variables.  To order a copy of the CART from CRC Press, click here.

Japanese Textbook on CART

Salford Systems' Dan Steinberg co-authored a Japanese book "Applied Tree Based Method by CART" with Yuji Horie and Atsushi Ootaki.   For more information and/or to order a copy of this book, click here.