Parametric Software Effort Estimation Based on Optimizing Correction Factors and Multiple Linear Regression

Context: Effort estimation is one of the essential phases that must be accurately predicted in early stage software project development. Currently, solving problems affect accuracy Use Case Points-based methods still a challenge to addressed. xmlns:xlink="http://www.w3.org/1999/xlink">Objective: This paper proposes parametric effort model based on Optimizing Correction Factors and Multiple Regression Models minimize error influence unsystematic noise, which has not been considered previous studies. The proposed method takes advantage Least Squared models Linear elements. xmlns:xlink="http://www.w3.org/1999/xlink">Method: We have conducted experimental research evaluate compare it with three related methods, i.e., 1) baseline – Points, 2) Factors, 3) Algorithmic Optimization Method. Experiments were performed datasets (Dataset D1, Dataset D2, D3). was analysed by applying various unbiased evaluation criteria statistical tests. xmlns:xlink="http://www.w3.org/1999/xlink">Results: results proved outperformed other improving accuracy. Statistically, significantly superior compared all tested datasets. xmlns:xlink="http://www.w3.org/1999/xlink">Conclusion: Based our obtained results, high capability helpful for managers during phase. correction factors are process.