Framework and Patterns for Machine Learning as Microservices Using Open Source Tools and Open Data
Abstract
Machine Learning is part of our everyday life, including social media analytics, online shopping, stocks performance prediction and other areas. Machine Learning is a complex process by itself. Explained at high level, takes data from around us and applies algorithms to it in order the generate intelligent outcomes using machine-based computation. There are multiple challenges with using and practicing Machine Learning, such as new technology adoption and handling constraints around accessing and using data, running computational workloads, using multiple cloud platforms, providers and proprietary technologies. This research explores new concepts and practical approach implementing and using machine learning in an easy, scalable, and sustainable way to solve common everyday problems. The research proposes adopting a machine learning framework defined using microservices architecture style, together with a series of patterns that provide practical guidance and are applicable for a broad set of machine learning tools and algorithms. Examples within this research refer to real estate market, specifically using open data available regarding single family real estate properties and real estate loans performance. Use cases and sample practices include real estate property data acquisition and real estate properties price prediction using machine learning patterns.
Subject Area
Computer science|Artificial intelligence|Computer Engineering
Recommended Citation
Barabasi, Istvan, "Framework and Patterns for Machine Learning as Microservices Using Open Source Tools and Open Data" (2019). ETD Collection for Pace University. AAI28091455.
https://digitalcommons.pace.edu/dissertations/AAI28091455
Remote User: Click Here to Login (must have Pace University remote login ID and password. Once logged in, click on the View More link above)