REGRESSION TREES
There have been a number of attempts to use regression trees in effort prediction [SRINIVASON & FISHER]. The results of which have shown that they can out perform proprietary models such as COCOMO, SLIM as so forth. Therefore they warrant a mention and should be taken seriously as an effort predication method. The best way to explain this method is to show an example of a regression tree illustrated below:
Click on the menu bar to change the dataset, and click on individual nodes to see the values.
This particular example predicts the testing time required for software development using particular tools.
A regression tree is created by examining what attributes can classify the data and then an algorithm constructs the tree, splitting nodes where appropriate. The example is very simple as the tree only ever splits into two leaf nodes, this is the most common format as it is easy to comprehend and check for logical errors.
Assessment of Regression Trees
