Abstract
Machine Learning deals with the issue of how to build programs that improve their performance at some task through experience. This paper deals with the subject of applying machine learning methods to software engineering. For effort estimation which not only provide an estimation but also confidence interval for it. The robust confidence intervals do not depend on the form of probability distribution of the errors in the training set. This paper compares various regression methods for software effort estimation with the help of number of experiments performed using NASA datasets and to show that robust confidence intervals can be successfully built.