Exploring machine learning techniques for software size estimation

Prediction models are fundamental in the early stages of the software development when many times, decisions must be taken without the required information. A typical information that is not available in these stages is software size metrics, such as lines of code (LOC). Models for LOC estimation are obtained from historical data and statistical regression methods are usually applied. These characteristics make this estimation problem especially interesting for the application of machine learning techniques. To explore this fact, this work applies Genetic Programming and Neural Networks techniques for LOC estimation. Two different data sets were used to obtain two models using respectively the metrics function points and number of components. The models are analysed and the machine learning techniques are compared.