Optimizing Machine Learning Model Accuracy via OBNT Algorithm: Advanced Data Preprocessing Technique

Machine learning has emerged as a focal point in the current landscape of data analytics. It involves the utilization of diverse ML algorithms to analyze data and extract valuable insights. These insights serve multiple industries, such as retail, petroleum, stock markets, and various organizations, resulting in increased sales, productivity, and profitability. Building an effective ML model to analyze and predict future data patterns is contingent upon high accuracy, which is directly dependent on the quality of the underlying data. Techniques like data preprocessing and feature selection play crucial roles in ensuring data quality. In this research, a methodology is proposed to meticulously select the most optimal data from raw datasets to construct an ML training model, thereby bolstering the accuracy percentage. The study involves comparing the accuracy of models built without employing the proposed technique against those utilizing it, alongside a comparative analysis of machine learning algorithm accuracy post-application of the proposed technique in contrast to existing preprocessing methodologies. The findings unequivocally demonstrate the efficacy of the proposed OBNT Algorithm in enhancing the accuracy of ML training models. The results indicated a statistically significant difference between the two groups, with a P value of 0.003 (p < 0.05).