Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/22708
Title: | Analisis Metode Clustering dengan Algoritma Density - Based Spatial Clustering of Application With Noise (DBSCAN) untuk Pengelompokan Cyberbullying di Instagram |
Other Titles: | Analysis of the Clustering Method with the Density - Based Spatial Clustering of Application With Noise (DBSCAN) Algorithm for Grouping Cyberbullying on Instagram |
Authors: | Sujatmiko, Budiman |
metadata.dc.contributor.advisor: | Lubis, Andre Hasudungan |
Keywords: | cyberbullying;clustering;data mining;DBSCAN;DBI |
Issue Date: | Sep-2023 |
Publisher: | Universitas Medan Area |
Series/Report no.: | NPM;198160018 |
Abstract: | Cyberbullying adalah bentuk dari sikap intimidasi yang dilakukan terhadap pengguna media sosial yang terjadi di dunia maya seperti Twitter, Facebook, Instagram. Sehingga, diperlukan adanya pengelompokan untuk mengatasi tindakan cyberbullying untuk membantu pihak – pihak berwenang seperti kominfo (Kementerian Komunikasi dan Informatika) untuk mengatasi dan menimalisir terjadinya tindakan cyberbullying yang terjadi di media sosial khususnya Instagram dan memberikan informasi edukasi tentang cyberbullying.Maka dari itu perlu adanya pengelompokan tindakan tersebut seperti penggunaan data mining untuk mengolah data yang besar dan membantu untuk mengambil keputusan. Pengelompokan pada data mining memiliki berbagai jenis algoritma, salah satunya adalah algoritma DBSCAN (Density-based spatial clustering of applications with noise). Algoritma DBSCAN adalah suatu metode pengelompokan berbasis kepadatan untuk melakukan clustering yang membangun area berdasarkan densitas yang terkoneksi (Setiawan dkk., 2021). Pada peneliti an ini, memilik jumla data 649 komentar yang berisi kalimat bullying dan nonbullying. Peneliti menggunakan algoritma DBSCAN sebagai metode untuk clustering kalimat cyberbullying dan menggunakan Davies Bouldin Index untuk mengevaluasi hasil pengelompokaan. Dari hasil perhitungan manual dilakukan dengan menentukan jarak Euclidean dan menghitung dengan algoritma DBSCAN dengan menentukan paramet epsilon adalah 0,4 dan nilai MinPts adalah 2 menghasilkan 3 kluster yaitu C1 {d1,d2,d3,d4,d5,d6}, C2 {d7,d8}, dan C3{d9,d10}. Dan hasil nilai dari perhitungan Davies Bouldin Index untuk klustering ini. Semakin rendah nilai DBI, maka semakin baik klusteringnya. Dalam perhitungan ini mendapatkan hasil sekitar 2,1826. Cyberbullying is a form of intimidation carried out against social media users that occurs in cyberspace such as Twitter, Facebook, Instagram. So, there is a need for a grouping to overcome cyberbullying acts to help authorities such as Kominfo (Ministry of Communication and Information Technology) to overcome and minimize the occurrence of cyberbullying acts that occur on social media, especially Instagram and provide educational information about cyberbullying. Therefore, it is necessary to group these actions, such as the use of data mining to process large data and help make decisions. Grouping in data mining has various types of algorithms, one of which is the DBSCAN (Density-based spatial clustering of applications with noise) algorithm. The DBSCAN algorithm is a density-based grouping method for clustering that builds areas based on connected density (Setiawan et al., 2021). In this research, the total data was 649 comments containing bullying and non-bullying sentences. Researchers used the DBSCAN algorithm as a method for clustering cyberbullying sentences and used the Davies Bouldin Index to evaluate the clustering results. From the results of manual calculations carried out by determining the Euclidean distance and calculating with the DBSCAN algorithm by determining the epsilon parameter is 0.4 and the MinPts value is 2 resulting in 3 clusters namely C1 {d1, d2, d3, d4, d5, d6}, C2 {d7, d8}, and C3{d9,d10}. And the value results from the Davies Bouldin Index calculation for this clustering. The lower the DBI value, the better the clustering. In this calculation the result is around 2.1826. |
Description: | 58 Halaman |
URI: | https://repositori.uma.ac.id/handle/123456789/22708 |
Appears in Collections: | SP - Informatic Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
198160018 - Budiman Sujatmiko Fulltext.pdf | Cover, Abstract, Chapter I, II, III, V, Bibliography | 2.03 MB | Adobe PDF | View/Open |
198160018 - Budiman Sujatmiko Chapter IV.pdf Restricted Access | Chapter IV | 677.34 kB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.