Data mining and spatial analysis of Twitter as a resource for assessing UK water pollution
Publication Date
December 14, 2023
Creator
Abstract
Water pollution is a significant global menace to human health, ecosystems, and economic progress. Despite advancements towards achieving the United Nations Sustainable Development Goals (SDGs) regarding accessible clean water and sanitation, water pollution remains a substantial hurdle. Active involvement of the public is vital in curbing water pollution, and comprehending their viewpoints and knowledge is pivotal for effective behavior modification. The indispensability of social media in our daily lives cannot be overstated, as it has emerged as the preeminent influential platform for educating society and gathering public perceptions. By utilizing data mining techniques, Social media platforms has the potential to convert public opinions into invaluable Volunteered Geographic Information (VGI), thus promoting citizen science. This study duly recognizes the potency of social
media, particularly Twitter, as an efficient tool for mapping water pollution patterns, trends, events, and public sentiments. By conducting spatial analysis and data mining on Twitter data, a wealth of valuable insights is unveiled. These insights encompass the identification of pollution hotspots, pinpointing event locations, the acquisition of knowledge regarding their underlying causes of pollution, regional disparities in these causes, measurement of sentiments on the topic across regions, and the formulation of potential resource management strategies. These contributions are in direct alignment with the overarching goal of achieving a net-zero-plus future, which resonates with the UN SDGs.
Item Type
ethesis
Thesis Type
MRes
Supervisors
Associated Schools / Departments
Faculty of Engineering
Department of Civil Engineering (UK)
eprints ID
76704
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
MRes_Thesis (20477598).pdf
Type
Full-text
Description
Examined. MRes Thesis
Size
5.19 MB
Format
Adobe PDF
Checksum (MD5)
bfd6ced3792b196218096a279f3bf0c5