IMPLEMENTASI STREAMLIT UNTUK MENGKLASIFIKASIKAN TINGKAT PENCEMARAN UDARA DI DKI JAKARTA DENGAN PENGGUNAAN ALGORITMA REGRESSION

SATRIA, HADY (2022) IMPLEMENTASI STREAMLIT UNTUK MENGKLASIFIKASIKAN TINGKAT PENCEMARAN UDARA DI DKI JAKARTA DENGAN PENGGUNAAN ALGORITMA REGRESSION. S1 thesis, Universitas Mercu Buana Jakarta-Menteng.

[img] Text (COVER)
COVER - HADY SATRIA.pdf
Download (3MB)
[img] Text (ABSTRAK)
ABSTRAK - HADY SATRIA.pdf
Download (24kB)
[img] Text (BAB 1)
BAB 1 - HADY SATRIA.pdf
Restricted to Registered users only
Download (109kB)
[img] Text (BAB 2)
BAB 2 - HADY SATRIA.pdf
Restricted to Registered users only
Download (93kB)
[img] Text (BAB 3)
BAB 3 - HADY SATRIA.pdf
Restricted to Registered users only
Download (59kB)
[img] Text (BAB 4)
BAB 4 - HADY SATRIA.pdf
Restricted to Registered users only
Download (93kB)
[img] Text (BAB 5)
BAB 5 - HADY SATRIA.pdf
Restricted to Registered users only
Download (409kB)
[img] Text (BAB 6)
BAB 6 - HADY SATRIA.pdf
Restricted to Registered users only
Download (283kB)
[img] Text (DAFTAR PUSTAKA)
DAFTAR PUSTAKA - HADY SATRIA.pdf
Restricted to Registered users only
Download (95kB)
[img] Text (LAMPIRAN)
LAMPIRAN - HADY SATRIA.pdf
Restricted to Registered users only
Download (275kB)
[img] Text (PERNYATAAN KEABSAHAN DAN PERSETUJUAN PUBLIKASI)
SCAN - HADY SATRIA.pdf
Restricted to Repository staff only
Download (395kB)

Abstract

Isu pencemaran udara cukup menjadi hal penting yang mengaitkannya dengan pertumbuhan hidup di perkotaan, isu ini biasa dikenal dengan polusi udara. Kondisi udara di Jakarta menunjukkan selama periode 2017-2020 umumnya udara dikategorikan tingkat sedang hingga tidak sehat. Hal ini diukur berdasarkan Indeks Standar Pencemaran Udara (ISPU) mengklasifikasikan kualitas udara di suatu daerah berdasarkan parameter zat-zat kimia serta akibat terhadap kesehatan. klasifikasi kualitas udara mulai dari kategori baik, sedang, tidak sehat, sangat tidak sehat, dan berbahaya, dapat diketahui dengan pemanfaatan algoritma tipe regression yaitu Logistic Regression (Regresi Logistik) dan Random Forest. Penelitian ini bermaksud untuk menghandalkan tipe algortima regression dengan merancang dashboard untuk mengklasifikasikan tingkat pencemaran udara di DKI Jakarta dengan library Steamlit. Selain itu, difungsikan sebagai validasi terhadap algortima yang telah di training terlebih dahulu mengenai tingkat evaluation dan keakuratannya. Dalam permasalahan klasifikasi indeks standar pencemar udara di DKI Jakarta menggunakan algoritma regression. Hasil evaluation modelling, untuk nilai sesuai berdasarkan proposi data split 80% : 20% dan 60% : 40% menujukkan bahwa secara accuracy lebih unggul algoritma Random Forest sebesar 96,66% & 96,22%. Dengan nilai MEA, MSE, RMSE terkecil secara berurut 0,03866, 0,04920, 0.2218. untuk data split 80% : 20%, kemudian nilai 0,045, 0,0615, 0,2480, pada data split 60% : 40%. Untuk nilai evaluasi r-square lebih besar yaitu 0.89910 & 0.87777 yang menunjukkan model yang dikeluarkan oleh regresi lebih baik. Dilakukan validitas menggunakan streamlit untuk menguji dataset terbaru dengan hasil menunjukkan variabel independent yang diinput dapat diklasifikasikan dengan hasil modeling algortima regression. The issue of air pollution is quite an important thing that relates to the growth of life in urban areas, this issue is commonly known as air pollution. Air conditions in Jakarta show that during the 2017-2020 period, the air is generally categorized as moderate to unhealthy. This is measured based on the Air Pollution Standard Index (ISPU) which classifies the air quality in an area based on the parameters of chemical substances and their effects on health. air quality classification ranging from good, moderate, unhealthy, very unhealthy, and dangerous categories, can be identified by using the regression type algorithm, namely Logistic Regression and Random Forest. This study intends to rely on the type of regression algorithm by designing a dashboard to classify the level of air pollution in DKI Jakarta using the Steamlit library. In addition, it functions as a validation of the algorithm that has been previously trained regarding the evaluation level and its accuracy. In the problem of classifying the air pollutant standard index in DKI Jakarta using a regression algorithm. The results of the modeling evaluation, for the appropriate value based on the proportion of split data 80%: 20% and 60%: 40%, show that the Random Forest algorithm is superior to the accuracy of 96.66% & 96.22%. With the smallest MEA, MSE, RMSE values, respectively, 0.03866, 0.04920, 0.2218. for split data 80% : 20%, then the value is 0.045, 0.0615, 0.2480, on split data 60% : 40%. For the evaluation value of r-square is greater, namely 0.89910 & 0.87777 which indicates the model issued by the regression is better. Validity using streamlit to test the latest dataset with results showing the inputted independent variables can be classified with the results of the regression algorithm modeling.

Item Type: Thesis (S1)
NIM/NIDN Creators: 41518110117
Uncontrolled Keywords: Algoritma Regression; Indeks Standar Pencemar Udara; Streamlit;Regression Algorithm; Air Pollutant Standard Index; Streamlit
Subjects: 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika
Divisions: Fakultas Ilmu Komputer > Informatika
Depositing User: Priyo Raharjo
Date Deposited: 09 Sep 2022 03:52
Last Modified: 09 Sep 2022 03:52
URI: http://repository.mercubuana.ac.id/id/eprint/68892

Actions (login required)

View Item View Item
UMB Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.