Public Decision-Making Analysis in Urban Heat Islands Based on Socioeconomic Modelling by Deep Learning Techniques

Autor: Prakash, S. Wilson, Kiruthigha, R., Yadav, Ajay Kumar, Kanna, S. K. Rajesh, Shaik, Khasimbee, Krishna, B. Murali
Zdroj: Remote Sensing in Earth Systems Sciences; 20240101, Issue: Preprints p1-10, 10p
Abstrakt: Urban heat island (UHI) effects, especially in highly urbanised areas, and greenhouse gas emissions from human activity are two elements that accelerate global climate change (GCC). Sustainable city planning and modification may benefit from the quantification of the surface UHI (SUHI) effect utilising land surface temperature (LST), local climate zones (LCZ), and deep learning methods like convolutional neural networks (CNN) and pix2pix. The findings show that socioeconomic variables account for 12–20% of variability in UHI intensity. In general, the contribution rate of urban economic scale is higher than that of population and industrial structure variables. The increase in urban economic expansion increased summertime heat discomfort. This research propose novel method in socioeconomic analysis based on urban heat pattern analysis with public healthcare system using decision-making with DL (deep learning) model. Input is collected as urban heat pattern dataset and processed for noise removal and normalisation. This data has been features are extracted and classified using convolutional fuzzy Q-Gaussian neural network and variational K-C-means regressive learning. The simulation analysis has been carried out based on training accuracy, average precision, recall, and F-measure. The proposed model attained average precision was 94%, recall was 90%, training accuracy was 97%, and F1-score was 92%.
Databáze: Supplemental Index
Externí odkaz: