Deep Artificial Neural Network for Software Effort Estimation

Accurate estimation is necessary for efficient project planning and resource allocation because software effort estimation is a crucial component of software development projects. The use of expert judgment in conventional software estimation techniques can be subject to error. In this study, we suggest a machine learning-based method for estimating software effort that makes use of predictive analysis to increase estimation precision. The suggested strategy entails gathering information from previous projects, such as project size, complexity, and development effort. We also take into account the development team’s performance, including their knowledge, proficiency, and productivity. Then, we put this information to use to train a machine learning model that can forecast how much work will be needed to finish new projects. In comparison to conventional estimation techniques, our approach has a number of benefits. With the aid of machine learning, we can take into account the complexity and variability of software development projects, pinpoint variables that could have an impact on project outcomes, and make better resource allocation and project planning decisions. The ANN algorithm has been applied in the current system. To improve accuracy, we have employed deep ANN in the proposed system. Artificial neural network is known as ANN. It is a kind of machine learning algorithm that is based on how the human brain works and is structured. Interconnected nodes, often referred to as neurons, are arranged in layers to make up an ANN. An ANN with numerous hidden layers is referred to as a “deep artificial neural network” (deep ANN). Deep ANNs can have many more hidden layers than standard ANNs, which enables them to learn more complicated data representations. Traditional ANNs typically have one or two hidden layers. The evaluation metrics are root mean squared error, mean absolute error, mean squared error, median magnitude relative error, and R-squared.