Automated Machine Learning: A Comprehensive Survey and Framework for Advancements

Abstract: Machine learning has become an indispensable tool in numerous domains, and the selection of an appropriate machine learning algorithm can significantly impact the success of a project. To address the challenge of algorithm selection, we present an Automated Machine Learning (AutoML) platform designed to empower users with the ability to effortlessly compare the performance of two selected machine learning algorithms on their datasets. This platform offers a user-friendly interface, guiding users through the entire process, from uploading their datasets to selecting algorithms and making informed decisions. The platform provides a streamlined workflow that allows users to easily upload datasets of their choice, select two machine learning algorithms from a pre-defined set, and compare their performance based on a variety of evaluation metrics. Users can visualize and analyze the results to gain insights into how different algorithms impact their data. Additionally, the platform offers assistance in identifying the most suitable algorithm for achieving optimal results. With the rapid evolution of machine learning techniques, this AutoML platform addresses the need for a user-friendly, efficient, and accessible tool to assist practitioners and researchers in algorithm selection. By simplifying the comparative analysis of machine learning models, this platform empowers users to make informed decisions that enhance the effectiveness of their data-driven projects.