PENERAPAN CLUSTERING PENGADUAN WARGA JAKARTA MENGGUNAKAN ALGORITMA FUZZY C-MEANS

Athoillah Sholahuddin, - (2023) PENERAPAN CLUSTERING PENGADUAN WARGA JAKARTA MENGGUNAKAN ALGORITMA FUZZY C-MEANS. S1 thesis, Universitas Pendidikan Indonesia.

	Text S_RPL_1905831_Title.pdf Download (625kB)
	Text S_RPL_1905831_Chapter1.pdf Download (122kB)
	Text S_RPL_1905831_Chapter2.pdf Restricted to Staf Perpustakaan Download (449kB)
	Text S_RPL_1905831_Chapter3.pdf Download (180kB)
	Text S_RPL_1905831_Chapter4.pdf Restricted to Staf Perpustakaan Download (1MB)
	Text S_RPL_1905831_Chapter5.pdf Download (107kB)
	Text S_RPL_1905831_Appendix.pdf Restricted to Staf Perpustakaan Download (346kB)

Abstract

Data pengaduan warga Jakarta sejak tahun 2017 sebanyak 931.578 laporan berbanding lurus dengan jumlah pertumbuhan penduduk mencerminkan banyak masalah yang dihadapi warga Jakarta sekaligus menjadi gambaran masalah di Jakarta. Data pengaduan tersebut tersedia dalam Open Data Jakarta, namun masih dalam bentuk mentah dan belum menjadi informasi. Penelitian ini bertujuan pemanfaatan data-data yang disediakan pemerintah menjadi ruang inovasi untuk memberikan informasi secara menyeluruh yang dapat dimanfaatkan warga Jakarta atas pengaduannya serta evaluasi pemerintah dalam mengatasi permasalahan dominan yang dihadapi warga melalui penerapan clustering menggunakan algoritma fuzzy c-means. Maka dari data pengaduan warga sebanyak 42.925 dikumpulkan dan diolah dengan preprocessing data. Penerapan metode fuzzy c-means menghasilkan 3 cluster terbaik berdasarkan nilai sum of square error, coefficient silhouette, dan calinski harabasz dengan masing-masing nilai 1997069,49; 1997069,49 dan 543288,98. Selanjutnya cluster evaluation didapat dengan nilai silhouette coefficient sebesar 0.79 atau 79%, nilai davies bouldin score sebesar 0,33, terakhir nilai calinski harabasz sebesar 543288,97 menunjukkan kualitas yang baik dalam penerapan clustering. Cluster-cluster tersebut membantu memahami karakteristik dan pola pengaduan warga Jakarta. Dalam hasil penelitian ini, ditemukan bahwa cluster 2 memiliki 24.521 jumlah data tertinggi dengan 39 kategori pengaduan, sementara cluster 0 dan 1 memiliki jumlah kategori pengaduan tertinggi dan terkecil yaitu 41 kategori dan 33 kategori. Dalam penelitian dilakukan proses deployment berupa sistem informasi yang dapat diakses oleh warga Jakarta sehingga menjadi sebuah informasi yang bermanfaat bagi warga dari pengaduan yang dilaporkan serta menjadi evaluasi kinerja pemerintah Jakarta. --------- Data on complaints from Jakarta residents since 2017 totaling 931,578 reports directly proportional to the number of population growth reflects the many problems faced by Jakarta residents as well as a picture of problems in Jakarta. The complaint data is available in Open Data Jakarta, but it is still in raw form and has not become information. This research aims to utilize the data provided by the government in an innovation space to provide comprehensive information that can be utilized by Jakarta residents for their complaints and evaluate the government's effectiveness in overcoming the dominant problems faced by residents through the application of clustering using the fuzzy c-means algorithm. From 42,925 citizen complaints collected and processed with data preprocessing The application of the fuzzy c-means method produces the 3 best clusters based on the sum of square error, coefficient silhouette, and Calinski harabasz values of 1997069.49, 1997069.49, and 543288.98, respectively. Furthermore, cluster evaluation is obtained with a silhouette coefficient value of 0.79, or 79%, a Davies-Bouldin score value of 0.33, and finally a Calinski-Harabasz value of 543288.9, showing good quality in the application of clustering. The clusters help to understand the characteristics and patterns of complaints from Jakarta residents. In the results of this study, it was found that cluster 2 has the highest amount of data (24,521) with 39 complaint categories, while clusters 0 and 1 have the highest and smallest number of complaint categories, namely 41 categories and 33 categories, respectively. In the research, the deployment process is carried out in the form of an information system that can be accessed by Jakarta residents so that it becomes useful information for residents from reported complaints and becomes an evaluation of the performance of the Jakarta government.

Item Type:	Thesis (S1)
Additional Information:	https://scholar.google.com/citations?view_op=new_profile&hl=en&authuser=4 SINTA ID : 6681751 SINTA ID : 6682222
Uncontrolled Keywords:	Open data Jakarta, Preprocessing data, Clustering, Fuzzy C-means, Cluster evaluation.
Subjects:	L Education > L Education (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General)
Divisions:	UPI Kampus cibiru > S1 Rekayasa Perangkaat Lunak
Depositing User:	Athoillah Sholahuddin
Date Deposited:	28 Aug 2023 04:07
Last Modified:	28 Aug 2023 04:07
URI:	http://repository.upi.edu/id/eprint/99304

Actions (login required)

View Item