Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/149017
Title: Design and development of machine learning technique for soil moisture sensor
Authors: Cheong, Yun Da
Keywords: Engineering::Computer science and engineering::Computing methodologies::Pattern recognition
Issue Date: 2021
Publisher: Nanyang Technological University
Source: Cheong, Y. D. (2021). Design and development of machine learning technique for soil moisture sensor. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/149017
Abstract: Soil Moisture is an important factor to plants, and it has been gaining a lot of attention in the agricultural sector. This study presents the design and development of Machine Learning technique, that aims to predict the soil moisture percentage using the total daily rainfall as an independent variable. Additional factors were added to the dataset to strive for the best results. These additional factors include three different locations in Singapore, three different timings of the day, as well as three different periods of the year. The three Machine Learning techniques used are Simple Linear Regression, Support Vector Regression and K-Nearest Neighbour. K-Nearest Neighbour produced the best results, and the different additional factors affected the predictability of the data. This study is conclusive that it is possible to predict the soil moisture data given the total daily rainfall, using ML methods. The average RMSE for all ML methods on all datasets is 4.01%, which is similar to the range of the soil moisture sensor used in this study.
URI: https://hdl.handle.net/10356/149017
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
CheongYunDa_U1721710D_Final_Report.pdf
  Restricted Access
2.48 MBAdobe PDFView/Open

Page view(s)

92
Updated on Jun 28, 2022

Download(s)

11
Updated on Jun 28, 2022

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.