CLASSIFICATION & REGRESSION TREES
CART®
As one of the most important and popular tools in modern data mining, CART is the ultimate classification tree. CART and its modeling engine have revolutionized the field of advanced analytics and inaugurated the current era of data science.
For those new to CART, it is a tree-based algorithm that works by looking at many various ways to locally partition or split data into smaller segments based on differing values and combinations of predictors. CART selects the best performing splits, then repeats this process recursively until the optimal collection is found. The result is a decision tree represented by a series of binary splits leading to terminal nodes that can be described by a set of specific rules. The tree and its layout is visually stimulating and intuitive to interpret so you don’t have to be a data scientist to understand and gain useful insights from it.
Designed for users of all levels, CART’s model can quickly reveal important relationships that could remain hidden when using other analytical tools. CART stands out in the predictive analytics field thanks to its original, highly desirable methodology that includes built-in automation, ease-of-use, performance, and accuracy.
Proprietary
CART’s methodology is based on a landmark mathematical theory introduced in 1984 by four world-renowned statisticians at Stanford University and the University of California at Berkeley. The CART modeling engine, Minitab’s implementation of Classification and Regression Trees, is the only decision tree software embodying the original proprietary code.
Fast & Versatile
Patented extensions to the CART modeling engine were specifically designed to enhance results for market research and analytics, support high-speed deployment, and predict and score in real time. Over the years, our engine has become one of the most popular, easy-to-use predictive modeling algorithms available and is fundamental to many modern data mining approaches based on bagging and boosting.
Interested in Deploying or Modeling Data with CART?
Minitab's Tree-Based Predictive Models
Whether you’re just getting started or looking to take your predictive analytics capabilities to the next level, Minitab’s tree-based modeling engines have the power you need.
CART® (Classification & Regression Trees)
The ultimate classification tree algorithm that revolutionized advanced analytics and inaugurated the current era of data science.
Random Forests®
The power to leverage multiple alternative analyses, randomization strategies, and ensemble learning in one convenient place.
TreeNet® (Gradient Boosting)
The most flexible and powerful machine learning tool that is capable of consistently generating extremely accurate models.