Automatic Machine Learning: An Exploratory Review
R. S. M. Lakshmi Patibandla, V. Sesha Srinivas, Sachi Nandan Mohanty, Chinmaya Ranjan Pattanaik · 2021
Automatic Machine Learning (AutoML) is also an exploratory region that has procured almost all complexes recently. However, the different methodologies followed by specialists and what has been unveiled by the accessible work is neither appropriately reported nor extremely clear because of the distinctions in the methodologies. While designing machine learning frameworks, AutoML serves as a bridge among different degrees of competence and aids the data science measure. Although a wide range of strategies is used to address this, there is no target link between these methods. AutoML is a start-to-finish strategy for automating the model development workflow without the need for external assistance. In the introduction session, this paper discussed the AutoML process, needs, difficulties, and benefits. AutoML methods, as well as Providers, have been explored inside the second session. Described Unified or AI Platform for different kinds for AutoML user in the third session. The fourth session is a case study of PayPal's e-commerce application, that analyzes both Automatic Machine Learning and Unified AutoML to get even more accurate results when using the unified AI platform, and the last session includes the conclusion.