Optimizing Linear Models via Sinusoidal Transformation for Boosted Machine Learning in Medicine

Sinusoidal Optimization of Linear Models

Abstract

Background: Machine learning relies on a hybrid of analytics, including regression analyses. There have been no attempts to deploy a sinusoidal transformation of data to enhance linear regression models.
Objectives: We aim to optimize linear models by implementing sinusoidal transformation to minimize the sum of squared error.
Methods: We implemented non-Bayesian statistics using SPSS and MatLab. We used Excel to generate 30 trials of linear regression models, and each has 1,000 observations. We utilized SPSS linear regression, Wilcoxon signed-rank test, and Cronbach’s alpha statistics to evaluate the performance of the optimization model. Results: The sinusoidal transformation succeeded by significantly reducing the sum of squared errors (P-value<0.001). Inter-item reliability testing confirmed the robust internal consistency of the model (Cronbach’s alpha=0.999). Conclusion: Our optimization model is valuable for high-impact research based on linear regression. It can reduce the computational processing demands for powerful real-time and predictive analytics of big data.

Keywords: