ANALISIS SEGMENTASI KONSUMEN MENGGUNAKAN ALGORITMA K-MEANS (Studi Kasus: Data Konsumen Mandiri Utama Finance)

Ashalea, Jingga Vradianti (2020) ANALISIS SEGMENTASI KONSUMEN MENGGUNAKAN ALGORITMA K-MEANS (Studi Kasus: Data Konsumen Mandiri Utama Finance). Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

Preview	Text ABSTRACT.pdf - Accepted Version Download (63kB) | Preview
Preview	Text ABSTRAK.pdf - Accepted Version Download (32kB) | Preview
Preview	Text COVER.pdf - Accepted Version Download (2MB) | Preview
Preview	Text BAB I.pdf - Accepted Version Download (165kB) | Preview
Preview	Text BAB II.pdf - Accepted Version Download (305kB) | Preview
Preview	Text BAB III.pdf - Accepted Version Download (234kB) | Preview
	Text BAB IV.pdf - Accepted Version Restricted to Registered users only Download (654kB)
Preview	Text BAB V.pdf - Accepted Version Download (35kB) | Preview
Preview	Text DAFTAR PUSTAKA.pdf - Accepted Version Download (110kB) | Preview
	Text LAMPIRAN.pdf - Accepted Version Restricted to Registered users only Download (693kB)

Abstract

Competition in the world's business is getting tighter, so every company must have a strategy that attracts consumers. The existing data sources have not been used by the company to be processed in determining the consumer segment using a computational approach. From the computational approach, there is no validation value which states that the K-means algorithm is suitable for analyzing consumer segmentation from the leasing company dataset. In the process, the method used is CRISP-DM which is included in six processes, namely business understanding, data understanding, data preparation, modeling, evaluation, and distribution. In the modeling stage, the algorithm used is K-means. K-Means can group data into several clusters based on data similarity. So that consumers who have the same segment will be grouped into one cluster and consumers who have different segments will be grouped in other clusters. This research will be based on DP and consumer loan interest. Meanwhile, the consumer segmentation studied includes the area, occupation and type of car chosen by the consumer. This study produces a clusterization output with a total of 3 clusters. Furthermore, the cluster formed by using the sillhoutte coefficient technique is tested and produces a value of 0.548 which means the cluster is good. Each cluster formed has a different result. So that the results can be a consideration for the company in determining the next promotion strategy. Keywords: CRISP-DM, Data mining, Clustering, K-means, consumers

Actions (login required)

View Item