This document analyzes and compares different software cost estimation techniques using machine learning algorithms. It uses the COCOMO and function point estimation models on NASA project datasets to test the performance of the ZeroR and M5Rules classifiers. The M5Rules classifier produced more accurate results with lower mean absolute errors and root mean squared errors compared to COCOMO, function points, and the ZeroR classifier. Therefore, the study suggests using M5Rules techniques to build models for more precise software effort estimation.