|
V13 N1 Paper 1
|
Annals of the MS in Computer Science and Information Systems at
UNC Wilmington
|
Spring 2019
|
Machine Learning for Internet of Things: Predict Refrigerator Performance Using Temperature Data
Dhaval Chauchan
Committee
Abstract
This paper discusses the architecture of Internet of Things and how Machine Learning can help to predict refrigerator temperature and performance. Internet of Things (IoT) is a term used for
a system which can collect data from various sources, such as lights, refrigerators, coolers, soil, water, air, etc. Sensors collect data and publish it to the cloud. This document explains methods
to use data and analyze it to predict the behavior of the system attached to the sensors. The sensors designed by Verisolutions Inc. Atlanta are used to collect refrigerator data. Temperature
and humidity data collected from the sensors is used for training the machine learning algorithm. As per the USDA guidelines, refrigerators should maintain a temperature of 40 °F or below. When
the refrigeration unit fails, the temperature inside rises in 3-4 hours. The machine learning algorithm can find a pattern and notify users with the potential failure alert.
download
(pdf)
Recommended Citation:
Chauchan, D., Ferner, C, Gebauer, J., Guinn, C. (2019) Machine Learning for Internet of Things: Predict Refrigerator Performance Using Temperature Data. Annals of the Master of Science in Computer Science and Information Systems at UNC Wilmington, 13(1) paper 1. http://csbapp.uncw.edu/data/mscsis/full.aspx.
|
V13 N1 Paper 1
|
Annals of the MS in Computer Science and Information Systems at
UNC Wilmington
|
Spring 2019
|
