Clustering Analysis of Bus Fares Trans Metro Deli Medan Using Mean Shif Clustering Method

Rambe, Rinanda Putri (2024) Clustering Analysis of Bus Fares Trans Metro Deli Medan Using Mean Shif Clustering Method. Journal of Computer System and Informatics (JoSYC), 5 (4). pp. 832-841.

Text 5728-Article_Text-22039-1-10-20240810.pdf Download (1MB)

Abstract

Medan City is the 3rd most populous city in Indonesia, according to data from the Central Statistics Agency, Medan has a population of 2.49 million in 2022, an increase from the previous 2.46 million in 2021. The increasing number of population inhabiting the city of Medan means that the need for transportation for the people of Medan is also increasing. Trans Metro Deli bus data can be grouped effectively using the mean shift algorithm based on several attributes, namely passenger category, payment method and fare. Each passenger group has different needs and ability to pay, which makes setting fair and efficient fares a challenge. Inappropriate pricing can lead to passenger dissatisfaction, reduce the number of public transportation users, and affect bus operators' revenue. Cluster technique is a well-known clustering technique, which aims to group data into clusters so that each cluster contains data that is as similar as possible. Mean shift belongs to the category of clustering algorithms with unsupervised learning that assigns data points to clusters iteratively by shifting the points towards the mode (mode is the highest density of data points in the region in the context of mean shift). Mean shift does not require determining the number of clusters in advance The attributes used in the clustering process, namely passenger category, payment method and fare can properly create a hyperplane between clusters, thus creating significant differences from each cluster, as evidenced by the silhouette score obtained by 0.64. By conducting this analysis, it is expected to find a more efficient and fair fare clustering pattern, and provide practical recommendations for management in setting fares that are more in line with passenger needs. In addition, this research also aims to evaluate the effectiveness of mean shift clustering in the context of transportation fare analysis.

Jenis Item:	Artikel
Subjects:	000 Generalities > 005 Computer programming, programs, data
Divisions:	Fakultas Sains dan Teknologi > Ilmu Komputer
Pengguna yang mendeposit:	Mrs Siti Masitah
Date Deposited:	03 Feb 2025 01:42
Last Modified:	03 Feb 2025 01:42
URI:	http://repository.uinsu.ac.id/id/eprint/24369

Actions (login required)

View Item