Big data water quality based on the IoT using LSTM and linear regression as an effort to achieve a Smart Green Campus

Kurniawan, Fachrul ORCID: https://orcid.org/0000-0002-3709-8764, Aziza, Miladina Rizka ORCID: https://orcid.org/0000-0001-8832-7398, Hasanah, Novrindah Alvi, Wibawa, Aji Prasetya, Hammad, Jehad and Alam, Laksamana Sulthan Big data water quality based on the IoT using LSTM and linear regression as an effort to achieve a Smart Green Campus. Research Report. Lembaga Penelitian dan Pengabdian kepada Masyarakat UIN Maulana Malik Ibrahim Malang. (Submitted)

Text 21288.pdf Restricted to Repository staff only Download (1MB)

Abstract

Combining big data analytics and water quality monitoring is essential for improving environmental sustainability, especially in smart green campuses. The notion of a “smart green campus” has gained increasing attention in recent years, as educational institutions seek to implement sustainable practices and reduce their environmental impacts, one of which is water quality. So, this study explored using big data and linear regression analysis to monitor and improve water quality. This work investigates using linear regression models to visualize and evaluate large datasets containing water quality characteristics such as pH, turbidity, temperature, and dissolved oxygen. The study uses big data collected from sensors across campus water sources to predict trends and detect abnormalities, allowing for more proactive environmental management. Visualizing these data trends provides campus managers with meaningful insights to optimize water consumption, improve filtration systems, and ensure the ecological health of campus water bodies. This technique not only enhances water conservation efforts but also contributes to the larger goal of developing a smart and sustainable campus ecology. This study highlights the potential of data-driven approaches to aid environmentally friendly campus operations and assist in the implementation of smart green campus programs. The results of this study demonstrate the successful integration of pH, turbidity, temperature and TDS sensors into IoT frameworks. Linear regression models were effectively applied for water quality classification. The model exhibited an accuracy and misclassification rate of 88.5% and 11.5%, respectively.

Item Type:	Research (Research Report)
Keywords:	big data; water monitoring; visualization; smart green campus
Subjects:	09 ENGINEERING > 0906 Electrical and Electronic Engineering > 090601 Circuits and Systems 09 ENGINEERING > 0906 Electrical and Electronic Engineering
Divisions:	Research and Community Service (Lembaga Penelitian dan Pengabdian kepada Masyarakat)
Depositing User:	Miladina Rizka Aziza
Date Deposited:	15 Nov 2024 10:17

Downloads

Origin of downloads

Actions (login required)

View Item