Software reusability metrics prediction and cost estimation by using machine learning algorithms

Abstract

In this research, a highly robust and efficient software design optimization model has been proposed for object-oriented programming based software solutions while considering the importance of quality and reliability. Due to a piece of information that software component reusability has allowed cost and time-efficient software design. The software reusability metrics prediction and cost estimation play a vital role in the software industry. Software quality prediction is an important feature that can be achieved a novel machine learning approach. It is a process of gathering and analyzing recurring patterns in software metrics. Machine learning techniques play a crucial role in intelligent decision making and proactive forecasting. This paper focuses on analyzing software reusability and cost estimation metrics by providing the data set. In the present world software, cost estimation and reusability prediction problem has been resolved using various newly developed methods. This paper emphasizes to solve the novel machine learning algorithms as well as improved Output layer self-connection recurrent neural networks (OLSRNN) with kernel fuzzy c-means clustering (KFCM). The investigational results confirmed the competence of the proposed method for solving software reusability and cost estimation.