Impact of Parallel Processing on LightGBM Implementation a Comparative Analysis CPU on Iris Plants Dataset

Abstract

Parallel processing has emerged as base for machine learning to address the computational requirements of complex models and expanded datasets. In additions, Parallel functions give the ability to algorithms exploiting the full potential of available accounting resources. This mechanism enhances parallel processing capabilities, as calculations are distributed through a multiple processor. This research explores the impact of the parallel processing of the central processing unit on the performance of LightGBM. The gradient-based learning in LightGBM enables efficient feature split decisions during tree construction. framework for scaling up, using the IRIS plant dataset. The study aims at comparing accuracy measures and training time for trained models with or without parallel processing units. The methodology includes advance data processing steps and the formation of environmentally sound management models with or without parallel processing units. The results reveal marked differences in accuracy and training time between the model and parallel processing of the central processing unit and its counterpart without it. Research contributes to understanding the role of parallel processing in the optimal functioning of the automated learning model.