Spectral Clustering Effect in Software Development Effort Estimation

Software development effort estimation is essential for software project planning and management. In this study, we present a spectral clustering algorithm based on symmetric matrixes as an option data processing. It expected that constructing model more similar can increase the accuracy. The research methods employ symmetrical processing experimentation. Four experimental models function point analysis, stepwise regression, clustering, categorical variables have been conducted. results indicate most advantageous variant combination of regression clustering. proposed method provides accurate estimates compared to baseline other tested variants.