Pontiselly, Pontiselly (2024) Monitoring Sentimen Ulasan Masyarakat di Google Maps Terhadap Layanan Kantor Imigrasi Surabaya Sebagai Early Warning System Untuk Perbaikan Kualitas Berbasis Convolutional Long Short Term Memory. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5003201083-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5003201083-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 July 2026. Download (8MB) | Request a copy |
Abstract
Perkembangan teknologi yang kian pesat di era globalisasi berdampak ke berbagai bidang, salah satunya pariwisata. Kemudahan dalam melakukan perpindahan tempat, khususnya negara, membuat makin meningkatnya jumlah wisatawan nasional ke mancanegara yang mana salah satu syaratnya adalah wajib memiliki paspor. Kolom komentar di Google Maps dapat menjadi wadah bagi masyarakat untuk memberikan feedback berupa ulasan terkait jasa layanan pembuatan paspor di kantor imigrasi. Data opini tersebut akan dianalisis sentimen guna membedakan antara ulasan kelas positif, netral, dan negatif. Sebelum itu, dilakukan pre-processing agar pembobotan kata menjadi vektor dapat dilakukan dengan Word2Vec. Lalu, analisis sentimen pada penelitian ini memanfaatkan metode hybrid, yaitu Convolutional Long Short Term Memory (Co-LSTM) yang dianggap mampu menghasilkan akurasi klasifikasi yang lebih tinggi daripada beberapa metode lainnya. Hasil analisis sentimen ini akan menjadi bahan untuk memonitor sentimen ulasan masyarakat yang mana opini kelas negatif diindikasikan sebagai produk cacatnya. Statistical Process Control (SPC) digunakan dalam upaya menyelesaikan masalah pengendalian kualitas. Metode SPC yang cocok untuk penelitian ini adalah peta kendali atribut p untuk memonitor layanan pembuatan paspor yang cacat. Hasil monitoring ini diharapkan dapat dijadikan insight bagi pihak institusi sebagai early warning system untuk melakukan evaluasi dan perbaikan terhadap layanan pembuatan paspor kedepannya. Data yang digunakan pada penelitian ini adalah ulasan masyarakat terkait layanan pembuatan paspor di kelima Kantor Imigrasi Kelas I Khusus TPI Surabaya yang di-scraping dari Google Maps bulan Desember 2018 hingga November 2023. Dari hasil analisis sentimen, didapatkan ulasan yang masuk kelas sentimen positif lebih banyak dibandingkan dengan sentimen negatif. Hasil ketepatan klasifikasi pada data training dan data testing didapatkan nilai AUC sebesar 98,61% (excellent classification) dan 85,66% (good classification). Hasil monitoring dengan peta kendali atribut p menggunakan data rating dan ulasan masyarakat memperlihatkan bahwa masih banyak pengamatan yang belum terkendali secara statistik yang berarti bahwa pihak Kantor Imigrasi Surabaya masih perlu untuk melakukan evaluasi dan perbaikan. Jenis kendala tertinggi yang dirasakan oleh masyarakat terkait layanan pembuatan paspor di kelima Kantor Imigrasi Surabaya adalah petugas yang kurang ramah dan sopan dalam melayani.
===================================================================================================================================
The rapid development of technology in this globalization era has a big impact has an impact on various fields, one of then is tourism. The ease of moving places, especially countries, has resulted in an increasing number of national tourists going abroad, where one of the conditions is that they must have a passport. The comments column on Google Maps can be a forum for the public to provide feedback in the form of reviews regarding passport production services at the immigration office. The opinion data will be analyzed for sentiment to differentiate between positive, neutral, and negative class reviews. Before that data will be pre-processing thus the word can be weighted into vectors with Word2Vec. Then, sentiment analysis in this research utilizes a hybrid method, namely Convolutional Long Short Term Memory (Co-LSTM) which is considered capable of producing higher classification accuracy than several other methods. The results of this sentiment analysis will be used as material for monitoring the sentiment of public reviews where negative class opinions are indicated as a defective product. Statistical Process Control (SPC) is used to resolve quality control problems. The SPC method that is suitable for this research is attribute control charts p to monitor parts of the application for defects. Hopefully, that the results of this analysis will be an insightful information used by the institution as an early warning system to evaluate and improve the passport making services in the future. The data used in this research are public reviews regarding passport making services at the Kantor Imigrasi Kelas I Khusus TPI Surabaya which were scraped from Google Maps from December 2018 until November 2023. From the results of sentiment analysis, it was found that more reviews were in the positive sentiment class compared to negative sentiment. The results of classification accuracy on training data and testing data obtained AUC values of 98.61% (excellent classification) and 85.66% (good classification). The results of monitoring with the p attribute control chart using rating data and public reviews show that there are still many observations that have not been statistically controlled, which means that the Kantor Imigrasi Surabaya still needs to carry out evaluations and improvements. The highest type of problem experienced by the public regarding passport making services at the Kantor Imigrasi Surabaya is officers who are not friendly and polite in their service.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Analisis Sentimen, Convolutional Long Short-Term Memory, Jasa Layanan, Paspor, Peta Kendali Atribut p; p Attribute Control Chart, M-Paspor, Sentiment Analysis |
| Subjects: | Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Pontiselly Pontiselly |
| Date Deposited: | 06 Feb 2024 06:03 |
| Last Modified: | 06 Feb 2024 06:03 |
| URI: | http://repository.its.ac.id/id/eprint/106258 |
Actions (login required)
 |
View Item |