Safira, Reza (2021) Implementasi Algoritma K-Means Dan Model Length, Recency, Frequency, Monetary (LRFM) Untuk Segmentasi Anggota (Studi Kasus: Koperasi XYZ). Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Text
05211740000133-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (5MB) | Request a copy |
Abstract
Koperasi merupakan salah satu bentuk organisasi yang memiliki peranan penting dalam meningkatkan perekonomian di Indonesia. Perkembangan koperasi di Jawa Timur, menurut data Kementrian Koperasi dan Usaha Kecil dan Menengah Republik Indonesia di tahun 2019 mencapai 21.757 unit koperasi yang aktif dan 12.089 unit koperasi yang memiliki sertifikat Nomor Induk Koperasi (NIK). Koperasi di sektor usaha jasa keuangan melayani jasa simpan pinjam. Koperasi XYZ merupakan salah satu koperasi yang melayani jasa simpanan dan pinjaman bagi anggotanya yang mengajukan. Oleh karena itu, anggota dianggap sebagai pelanggan. Dalam menjalankan jasa pinjaman, koperasi diharapkan mendapatkan pendapatan yang disebut juga dengan Sisa Hasil Usaha (SHU) dari uang jasa yang dibayarkan oleh anggota. SHU juga nantinya akan dibagi kepada setiap anggota sesuai dengan partisipasinya. Partisipasi tersebut dapat dilihat dari keaktifan anggota dalam menggunakan jasa pinjaman koperasi XYZ. Pada tahun 2018-2020 terjadi penurunan partisipasi anggota terhadap jasa pinjaman, disebabkan kurangnya strategi penawaran pihak koperasi. Untuk menyelesaikan permasalahan tersebut salah satu solusi yang dapat dilakukan yaitu melakukan segmentasi anggota. Mengelompokkan perilaku anggota dalam melakukan transaksi pinjaman dengan menerapkan ekstraksi fitur LRFM. Kemudian, segmentasi yang dilakukan menggunakan metode K-Means, dengan penentuan jumlah K menggunakan metode elbow. Selanjutnya, untuk melakukan validasi jumlah Cluster menggunakan metode Silhouette dan DBI. Hasil dari pengerjaan tugas akhir ini menggunakan fitur LRFM, metode K-Means, dan pembobotan AHP menghasilkan 3 Cluster anggota. Hasil segmentasi ditampilkan dalam bentuk visualisasi untuk memudahkan dalam melakukan analisis setiap Cluster yang terbentuk sesuai karakteristiknya. Karakteristik yang terbentuk tidak terlihat perbedaan yang signifikan, karena data yang tersedia acak atau tidak berpola. Namun, hasil cluster dilakukan pemetaan dalam customer value matrix dan customer relationship matrix, yang nantinya dapat dijadikan acuan dalam penyusunan rekomendasi strategi.
====================================================================================================
Cooperatives are one form of organization that has an important role in improving the economy in Indonesia. According to data from the Ministry of Cooperatives and Small and Medium Enterprises of the Republic of Indonesia, the development of cooperatives in East Java in 2019 reached 21,757 active cooperative units and 12,089 cooperative units that have a Cooperative Identification Number (NIK) certificate. Cooperatives in the financial services business sector serve savings and loan services. XYZ Cooperative is one of the cooperatives that provide savings and loan services for its members who apply. Therefore, members are considered as customers. In carrying out loan services, cooperatives are expected to get income which is also known as Remaining Operating Results (SHU) from service fees paid by members. SHU will also be distributed to each member according to their participation. This participation can be seen from the activeness of members in using XYZ cooperative loan services. In 2018-2020, member participation in loan services decreased due to the lack of a cooperative offering strategy. To solve these problems, one solution that can be done is segmenting members. Grouping the behavior of members in conducting loan transactions by implementing LRFM feature extraction.Then, segmentation is done using the K-Means method, with the determination of the number of K using the elbow method. Furthermore, to validate the number of clusters using the Silhouette and DBI methods. The results of this final project using the LRFM feature, K-Means method, and AHP weighting produce 3 Cluster members. Segmentation results are displayed in the form of visualization to facilitate the analysis of each cluster that is formed according to its characteristics. The characteristics formed did not show a significant difference, because the available data were random or not patterned. However, the results of the cluster are mapped into the customer value matrix and customer relationship matrix, which can later be used as a reference in formulating strategic recommendations
Item Type: | Thesis (Undergraduate) |
---|---|
Uncontrolled Keywords: | Segmentasi Pelanggan, Clustering, K-Means, Fitur LRFM, Analytical Hierarchy Process. |
Subjects: | H Social Sciences > HF Commerce > HF5415.5 Customer services. Customer relations Q Science > QA Mathematics > QA278.55 Cluster analysis |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis |
Depositing User: | Reza Safira |
Date Deposited: | 23 Aug 2021 10:20 |
Last Modified: | 23 Aug 2021 10:20 |
URI: | http://repository.its.ac.id/id/eprint/88932 |
Actions (login required)
View Item |