ENHANCING THE PROCESS EFFICIENCY OF RPA USING MACHINE LEARNING

Abstract

The robotic process automation (RPA) has completely changed commercial processes by automating repetitive procedures and reducing human error. This study investigates how machine learning (ML) can be integrated into RPA to improve it. The paper talks about the challenges of picking the right processes for automation and exposes the advantages of RPA, like the ability to imitate human-computer interactions. It argues that decision-making, predictive maintenance, and anomaly detection can be done better by RPA systems when techniques including decision trees as well as random forests from ML are used. According to the study, it is evident that the classification algorithms with the highest accuracy were the Gradient Boosting Machine (GBM) and Random Forest Classifier (RCF) that is 98%, while the others had a lower accuracy of Decision Forest Classifier (DCF) 97%, Random Regression Forest (RRF) 93%, and lastly the Decision Random Forest (DRF) 87%. Machine learning through sophisticated technologies such as optical character recognition and natural language processing improves data extraction, processing accuracy, and user engagement. According to the findings, combining RPA with ML leads to an automated system that is more flexible, resilient and intelligent which can greatly enhance operations in a variety of industries. This comprehensive approach taken by this research will be useful for both academicians and practitioners in developing and optimizing RPA systems, finally advancing intelligent automation.