An Optimized Analogy-Based Project Effort Estimation
This work addresses a specific issue in software engineering estimation, but it is incremental as it builds on existing ABE methods.
The paper tackled the problem of predicting the optimal number of analogies and adjustment techniques in Analogy-Based Estimation for project effort, resulting in a new model that improved predictive performance and showed variable analogy counts per project.
Despite the predictive performance of Analogy-Based Estimation (ABE) in generating better effort estimates, there is no consensus on how to predict the best number of analogies, and which adjustment technique produces better estimates. This paper proposes a new adjusted ABE model based on optimizing and approximating complex relationships between features and reflects that approximation on the final estimate. The results show that the predictive performance of ABE has noticeably been improved, and the number of analogies was remarkably variable for each test project.