Software testing effort estimation with machine learning

One of the main headlines of software test literature is the problem of not having a sound estimation of software test effort while scheduling a plan for the whole software development. The software test process time in software projects shouldbe estimated timely in order to gather the required resources beforehand. In this work, using Machine Learning algorithms, we propose a new method of software effort estimation. Using the past experiences of software test efforts processed in ASELSAN in the areas of command center graphical user interfaces, embedded and real-time software test developments, we strive for better estimations. The new estimations are evaluated in comparison with the traditional methods.