STRING (Satuan Tulisan Riset dan Inovasi Teknologi)
Vol 8, No 1 (2023)

Perbandingan Evaluasi Metode Davies Bouldin, Elbow dan Silhouette pada Model Clustering dengan Menggunakan Algoritma K-Means

Muhammad Sholeh (Unknown)
Khurotul Aeni (Universitas Peradaban)



Article Info

Publish Date
05 Aug 2023

Abstract

One of data mining model that is often used is the clustering model. The clustering model is used to create a grouping of a datasheet. Data clustering can be used to distinguish data in a datasheet.  Many of the best groupings can be done with the clustering model evaluation process. In the clustering method, the evaluation process can use various evaluation methods. The research will cluster the results of a survey on reviews of tourist destinations consisting of 10 categories. The method used is a data mining method, namely Knowledge Discovery in Database (KDD). Stages in KDD include data selection, data cleaning, transformation, data mining process and evaluation of model results. The process of creating a clustering model uses a public datasheet, namely tripadvisor_review.csv.  The data clustering process uses the K-means algorithm. The result of the clustering will be tested by comparing evaluation methods. This evaluation method is used to select the best amount of clustering. Testing to get the best clustering results is conducted by testing from clustering 2 to 15. The evaluation uses Davies Bouldin, Elbow and Silhouette methods. The result shows that the number of datasheet groupings with the three evaluation methods provides recommendations for grouping as many as 2 groups. 

Copyrights © 2023






Journal Info

Abbrev

STRING

Publisher

Subject

Computer Science & IT Mathematics

Description

STRING (Satuan Tulisan Riset dan Inovasi Teknologi) focuses on the publication of the results of scientific research related to the science and technology. STRING publishes scholarly articles in Science and Technology Focus and Scope Covering: 1. Computing and Informatics 2. Industrial Engineering ...