Modeling Coastal Area Change Analysis of Coastal Urban Areas at Semarang City, Indonesia

Modeling Coastal Area Change Analysis of Coastal Urban Areas at Semarang City, Indonesia: A Comparison of Machine Learning Classifiers on Optical Satellite Imageries Data

Authors

  • Renata De La Rosa Manik Politeknik Statistika STIS
  • Arie Wahyu Wijayanto Politeknik Statistika STIS

DOI:

https://doi.org/10.34123/icdsos.v2023i1.367

Keywords:

Coastal, Changes, Semarang City, Machine Learning, Optical Satellite

Abstract

A coastal area is defined as the boundary between land and sea. Coastal urban areas are susceptible to various hazards that are becoming more severe, such as flooding, erosion, and subsidence due to a mix of man-made and natural factors, including urbanization and climate change. Regardless of the high importance of coastal area monitoring, conducting field surveys is expensive, time-consuming, and geographically limited to non-remote regions. Semarang City is one of the cities in Indonesia that is at risk of changes in its coastline and causes various natural problems. This research aims to estimate changes in the coastal land area in Semarang City. In observing the phenomenon of changes in area in coastal areas in Semarang City, remote sensing technology with Sentinel-2 satellite imagery was used. This research implements and compares the Random Forest (RF) and Support Vector Machine (SVM) machine learning methods in building classification models. From the results of land area in 2019, 2021, and 2023 with the best classification model, namely SVM, information was obtained on an increase in coastal area of 387.94 ha in 2021, then a change in area decrease of 417.32 ha in 2023.

Downloads

Published

2023-12-29

How to Cite

Manik, R. D. L. R., & Wijayanto, A. W. (2023). Modeling Coastal Area Change Analysis of Coastal Urban Areas at Semarang City, Indonesia: Modeling Coastal Area Change Analysis of Coastal Urban Areas at Semarang City, Indonesia: A Comparison of Machine Learning Classifiers on Optical Satellite Imageries Data. Proceedings of The International Conference on Data Science and Official Statistics, 2023(1), 265–273. https://doi.org/10.34123/icdsos.v2023i1.367