Development of Effective Artificial Neural Network Model using Sequential Sensitivity Analysis and Randomized Training

As the machine learning algorithms evolve, there is a growing need of how to train algorithm effectively for large data with available resources in practically less time. The paper presents an idea developing effective model that focuses on implementation sequential sensitivity analysis and randomized training approach which can be one solution this need. Many researchers focused eliminate insignificant features ands reduce complexity selection. These methods relatively take time validation through modeling hence found impractical data. On other hand, was most popular but very brief explanation research articles method meaningful getting higher accuracy. current work use artificial neural network (ANN) high dimensionality thermal power plant (SSA) technique includes correlation (CA), Analysis variance (ANOVA), Akaike information criterion (AIC) manner all possible feature combinations. Only selected combinations are then tested against different such as downward extrapolation, upward interpolation ANN. also suggesting significance comparison-based qualitative reasoning. statistical parameters, mean square error (RMSE), Mean absolute relative difference (MARD) R Square (R^2)were accessed purposes. mainly useful field Ecommerce, Finance, industry facilities where generated.