Wijanarko, Adi Pradipta Satya (2019) Metode Two-Step Cluster Pada Preferensi Mahasiswa Dalam Berbelanja Online (Studi pada Mahasiswa FMIPA Universitas Brawijaya Malang). Sarjana thesis, Universitas Brawijaya.
Abstract
Analisis cluster merupakan teknik multivariat yang digunakan untuk mengelompokkan objek menjadi beberapa kelompok berdasarkan ukuran kemiripan yang diterapkan pada data kontinu, kategorik atau campuran antara data kontinu dan kategorik. Metode yang diterapkan pada data campuran salah satunya adalah Two-Step Cluster Method (TSCM). Mahasiswa dalam universitas, fakultas maupun program studi memiliki prefrensi atau kebutuhan yang berbeda dalam belanja online. Penelitian akan menerapkan analisis TSCM pada karakteristik mahasiswa ditinjau berdasarkan enam peubah, yaitu: jenis kelamin, platform e-commerce, jenis pembayaran, jenis barang, biaya belanja online per bulan dan uang saku per bulan. Tujuan penelitian adalah menerapkan TSCM pada data mahasiswa S1 FMIPA Universitas Brawijaya agar diperoleh cluster yang beranggotakan mahasiswa sesuai karakteristik. Pengelompokan tahap awal menerapkan metode sekuensial dan tahap akhir menerapkan metode pemusatan berdasarkan ukuran jarak Log- Likelihood. Penentuan banyak cluster optimal didasarkan pada nilai Bayesian Information Criterion (BIC). Hasil analisis Two-Step Cluster menunjukkan bahwa dari 97 mahasiswa terdapat dua cluster optimal yang terbentuk. Cluster pertama terdapat 51 mahasiswa dan cluster kedua terdapat 46 mahasiswa. Cluster 1 bernama cluster spesifik karena hanya terdapat mahasiswa perempuan dan Shopee sebagai platform e-commerce pilihan, sementara cluster 2 bernama cluster general karena terdapat mahasiswa laki-lak maupun mahasiswa perempuan dan pilihan platform e-commerce yang lebih variatif
English Abstract
Cluster analysis is a multivariate technique that is used to group objects into several clusters based on similarity measure, applied to continuous, categorical or mixed data (continuous and categorical). Two Step Cluster Method (TSCM) method can be applied to the mixed data. Students in the university, faculty or courses have different characteristics, so TSCM method can be applied to data about the characteristics of college students. The research will using TSCM method on student characteristics based on six variables: gender, platform e-commerce, payment method, type of item, monthly pocket money and budget online shop. The purpose is to applying TSC method on data of college students, to get cluster consist of students based on characteristics. Pre-clustering applying the sequential method and clustering of cases applying agglomerative method based on measurements of Log-Likelihood distance. The determination of the number of optimal clusters based on the value of the Bayesian Information Criterion (BIC). The result of analysis shows that from 97 college students, two optimal clusters are formed. The first cluster consists of 51 student, while the second cluster consists of 46 student. Cluster one named specific cluster, because only have a woman student and shopee for a platform e-commerce choice, then cluster two named general cluster becase have a man and woman student and have more variative platform e-commerce choice.
Other obstract
-
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/MIPA/2019/44/051910746 |
| Uncontrolled Keywords: | Analisis Cluster, Belanja Online Mahasiswa, Jarak Log-Likelihood, Two Step Cluster Method (TSCM). Cluster Analysis, Log-Likelihood Distance, Student Online Shop Two-Step Cluster (TSCM). |
| Subjects: | 500 Natural sciences and mathematics > 519 Probabilities and applied mathematics > 519.5 Statistical mathematics > 519.53 Descriptive statistics, multivariate analysis, analysis of variance and covariance |
| Divisions: | Fakultas Matematika dan Ilmu Pengetahuan Alam > Statistika |
| Depositing User: | Budi Wahyono Wahyono |
| Date Deposited: | 24 Aug 2020 07:33 |
| Last Modified: | 29 Mar 2022 03:45 |
| URI: | http://repository.ub.ac.id/id/eprint/176769 |
Preview |
Text
Adi Pradipta Satya Wijanarko.pdf Download (2MB) | Preview |
Actions (login required)
 |
View Item |