Visual representation of predictions in software development based on software metrics history data
Blažo Popović, Adis Balota, Dzenan Strujic · 2016
Software who is being developed for more than a year, is categorized in the group of large projects, and is entering the area of critical and risky in terms of their successful completion. Because of those reasons, they require constant monitoring by the software manager. In order to perform adequate monitoring of developed solution and its projects, there are a number of software metrics that provide these information in numerical form, but very few of them is displayed to software manager in visual form. Predictions for future solution development, and its visual representation, based on historical data gathered from the same system, are usually not included in such systems. These kind of systems must have the ability to predict state of the developed system in the future, and adapt its output upon having analyzed his software metrics history data. In this paper, we propose a intelligent system that will analyze software metrics history data of the solution, and make predictions and visual outputs to support software manager decisions. The results show that using this decision support tool, as a form of intelligent system, helps software managers in their decision making during project management, and in reducing overall project risk.