IMPLEMENTASI MODEL DEEP LEARNING CNN-LSTM UNTUK PRAKIRAAN KUALITAS UDARA BERDASARKAN KADAR POLUTAN DI KOTA JAKARTA

Yaasintha La Jopin Arisca Corpputy, . (2024) IMPLEMENTASI MODEL DEEP LEARNING CNN-LSTM UNTUK PRAKIRAAN KUALITAS UDARA BERDASARKAN KADAR POLUTAN DI KOTA JAKARTA. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.

	Text ABSTRAK.pdf Download (133kB)
	Text AWAL.pdf Download (1MB)
	Text BAB I.pdf Download (174kB)
	Text BAB II.pdf Restricted to Repository UPNVJ Only Download (1MB)
	Text BAB III.pdf Restricted to Repository UPNVJ Only Download (1MB)
	Text BAB IV.pdf Restricted to Repository UPNVJ Only Download (2MB)
	Text BAB V.pdf Download (162kB)
	Text DAFTAR PUSTAKA.pdf Download (183kB)
	Text RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (368kB)
	Text LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (322kB)
	Text HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (312kB)
	Text ARTIKEL KI.pdf Restricted to Repository staff only Download (1MB)

Abstract

The problem of air pollution due to declining air quality in big cities like Jakarta has become a crucial issue. Data from aqicn.org shows that by 2023, Jakarta's Air Quality Index (AQI) will worsen to 154, which falls into the “unhealthy” category. This condition has a significant impact, not only on public health, but also on the environment and economy. Therefore, forecasting polluted air quality is important as an instrument for the public and policy makers in future air quality management efforts. This research aims to implement Deep Learning models based on Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) to forecast air quality based on pollutant levels in Jakarta City. The dataset used includes the Jakarta Air Pollution Standard Index (ISPU) from 2021 to 2024. Model evaluation is performed using the Root Mean Squared Error (RMSE) metric with a series of trials and parameter tuning, such as the number of convolution filters, LSTM hidden layer, window size, and batch size, to achieve the most optimal performance. The results show that the CNN-LSTM model with parameter tuning produces the best RMSE value of 0.059238. This model also shows consistent evaluation results at various measurement stations and has the best generalization ability in predicting air quality parameter at the location or station DKI4 (Lubang Buaya). In addition, this model also shows the highest accuracy in predicting SO2 and O3 pollutant concentrations with RMSE values of 1.733 and 3.182, respectively.

Item Type:	Thesis (Skripsi)
Additional Information:	[No.Panggil : 2010511091] [Pembimbing 1 : Neny Rosmawarni] [Pembimbing 2 : Novi Trisman Hadi] [Penguji 1 : Bayu Hananto] [Penguji 2 : Hamonangan Kinantan P.]
Uncontrolled Keywords:	Air Quality Forecast, CNN-LSTM, Deep Learning, ISPU, Pollutants
Subjects:	Q Science > Q Science (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software
Divisions:	Fakultas Ilmu Komputer > Program Studi Informatika (S1)
Depositing User:	YAASINTHA LA JOPIN ARISCA CORPPUTY
Date Deposited:	04 Feb 2025 08:37
Last Modified:	04 Feb 2025 08:37
URI:	http://repository.upnvj.ac.id/id/eprint/35774

Actions (login required)

View Item