Implementasi Text Mining Dalam Mengevaluasi Inovasi Layanan Dispendukcapil Surabaya Berdasarkan Data Keluhan Pemohon

Indriasari, Irnanda Dwi Ayu (2021) Implementasi Text Mining Dalam Mengevaluasi Inovasi Layanan Dispendukcapil Surabaya Berdasarkan Data Keluhan Pemohon. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 02411740000073-Undergraduate_Thesis.pdf] Text
02411740000073-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (3MB) | Request a copy
[thumbnail of 02411740000073-Undergraduate_Thesis.pdf] Text
02411740000073-Undergraduate_Thesis.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy

Abstract

Kebutuhan administrasi kependudukan dan pencatatan sipil warga Surabaya adalah salah satu jenis layanan publik yang difasilitasi oleh Dispendukcapil Surabaya. Inovasi mulai dari masa SIAK (Sistem Integrasi Administrasi Kependudukan) di mana seluruh database kependudukan masyarakat Indonesia ada dalam satu sistem hingga masa KLAMPID (Kawin, Lahir, Mati, Pindah, dan Datang) di mana kurang lebih 77 layanan saat ini difasilitasi secara online. Untuk mengetahui apakah inovasi layanan Dispendukcapil sudah menjawab kebutuhan warga, dilakukan pengolahan pada data keluhan yang bersumber dari service desk warga, service desk telepon, call center, dan rekapitulasi keluhan masyarakat. Pengolahan data statistik deskriptif sederhana memiliki beberapa keterbatasan seperti hasil belum tentu merepresentasikan seluruh keluhan dan pencarian kategori dilakukan secara manual. Maka dari itu, metode yang digunakan dalam penelitian ini adalah text mining yang terbagi menjadi proses pre-processing, TF-IDF vectoring, dan K-Means clustering yang pada akhirnya divisualisasikan dengan word cloud. Banyak klaster yang diperoleh sebanyak 15 didapatkan melalui pertimbangan nilai coefficient silhoutte dan klaster yang dihasilkan. Hasil word cloud menunjukkan bahwa terdapat beberapa klaster masih menghasilkan kesamaan antar klasternya dan kurang optimalnya pengolahan karena kualitas data keluhan serta tingginya variasi data. Hasil word cloud menampilkan bahwa pada masa KLAMPID, masalah yang paling sering dikeluhkan berkaitan dengan pengajuan online, keterlibatan pihak eksternal, dan mekanisme foto pada pembuatan akun KLAMPID yang akan difokuskan menjadi rekomendasi perbaikan.

=========================================================
The need for population administration and civil registration for the citizens of Surabaya is one type of public service that is facilitated by the Dispendukcapil Surabaya. The innovations started from the SIAK period (System Integration of Population Administration) where all Indonesian population databases were in one system to the KLAMPID period (Marriage, Birth, Death, Move, and Coming) where approximately 77 services are currently facilitated online. To find out whether the Dispendukcapil service innovation has answered the residents' needs, processing of complaint data is carried out from the citizen service desk, telephone service desk, call center, and recapitulation of community complaints. Simple descriptive statistical data processing has some limitations such as the results do not necessarily represent all complaints and the search for categories is done manually. Therefore, the method used in this research is text mining which is divided into pre-processing, TF-IDF vectoring, and K-Means clustering which are finally visualized with a word cloud. The number of clusters obtained is 15 obtained through consideration of the coefficient silhoutte’s value and the result of the clusters. The results of the word cloud show that there are several clusters that still produce similarities between their clusters and less than optimal processing due to the quality of the complaint data and the high variation of the data. The results of the word cloud show that during the KLAMPID period, the problems most frequently complained about were related to online submissions, the involvement of external parties, and the photo mechanism in creating a KLAMPID account which would be focused on recommendations for improvement.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: layanan publik, KLAMPID, kepuasan masyarakat, text mining, K-Means Clustering public services, KLAMPID, customer satisfaction, text mining, K-Means Clustering
Subjects: T Technology > T Technology (General) > T57.5 Data Processing
Divisions: Faculty of Industrial Technology and Systems Engineering (INDSYS) > Industrial Engineering > 26201-(S1) Undergraduate Thesis
Depositing User: Irnanda Dwi Ayu Indriasari
Date Deposited: 20 Aug 2021 02:39
Last Modified: 20 Aug 2021 02:39
URI: http://repository.its.ac.id/id/eprint/87872

Actions (login required)

View Item View Item