The optimization of reconfigured real-time datasets for improving classification performance of machine learning algorithms
Abstract
Machine Learning (ML) is one of the modern techniques with promising outcomes in classification and prediction domains. In order to improve demand projections, historical data can be analyzed with different methods, such as ML techniques, time series analysis, and deep learning models. A knowledgeable demand visualization system is developed in this work. This enhanced model is based on the study investigations and interpretation of statistical data using various prediction methods, including traditional time series techniques, support~vector machines, and deep learning algorithms. Therefore, we are aiming to analyze the effectiveness of different model classification for ML to prevent the use of personalized log
data. We offer classification pre-processing steps to improve the performance results and speed of the eXtended Machine Learning (X-ML) algorithm in the training data sets. We also proved~the classification speed improvement on several real datasets via this automated system. The improvement is significantly greater and coherent, with the accuracy dropped in only a few case scenarios.