Aditya, Adiella Fadia (2024) Deep Embedded K-Medoids pada Data Ulasan Aplikasi Gojek yang Diambil dari Google Play Store. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
5002201013-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (8MB) | Request a copy |
Abstract
Untuk mempertahankan dan meningkatkan mutu pelayanan, PT Gojek Indonesia membutuhkan informasi dari ulasan pelanggan. Ulasan pelanggan tersebut ada yang berupa teks dan dapat diolah dengan beberapa teknik, salah satunya pengelompokan teks (text clustering). Penelitian ini membahas mengenai aplikasi penerapan Algoritma Deep Embedded K-Medoids (DEKM) untuk mengelompokan data ulasan pengguna sebagai informasi umpan balik terhadap aplikasi. Algoritma ini dilakukan dengan transformasi lanjutan di ruang embedding untuk memperoleh informasi struktur cluster melalui matriks transformasi ortonormal. Data yang digunakan adalah kumpulan data teks ulasan aplikasi Gojek yang diperoleh melalui teknik web scraping pada website Google Play Store. Dari hasil eksperimen, ditemukan bahwa Algoritma DEKM belum mampu memberikan hasil akurasi yang tinggi yaitu 42,95% disebabkan kualitas fitur dan target pada dataset belum bisa secara akurat merepresentasikan informasi yang ada pada data asli.
==============================================================
To maintain and improve service quality, PT Gojek Indonesia needs information from customer reviews. Some of these customer reviews are in the form of text and can be processed using several techniques, one of which is text clustering. This research discusses the application of the Deep Embedded K-Medoids (DEKM) algorithm to group user review data as feedback information for applications. The data used is a collection of Gojek application review text data obtained through web scraping techniques on the Google Play Store website. From the experimental results, it was found that the DEKM algorithm has not been able to provide high accuracy results, namely 42,95%, because the quality of the features and targets in the dataset has not been able to accurately represent the information in the original data.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | Clustering, Deep Embedded K-Medoids, Ulasan Gojek, Clustering, Deep Embedded K-Medoids, Gojek Review |
Subjects: | Q Science > QA Mathematics > QA278.55 Cluster analysis Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) |
Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis |
Depositing User: | Adiella Fadia Aditya |
Date Deposited: | 07 Aug 2024 01:00 |
Last Modified: | 07 Aug 2024 01:00 |
URI: | http://repository.its.ac.id/id/eprint/114412 |
Actions (login required)
View Item |