University of Twente Student Theses

Login

Comparison of different types of cluster algorithm on Netherland telecommunication provider solutions

Wicaksana, J. (2021) Comparison of different types of cluster algorithm on Netherland telecommunication provider solutions.

[img] PDF
893kB
Abstract:Clustering is an unsupervised method of grouping data into different groups. In this research the data is in a form of text. The text then needs to be translated into numbers with word2vec skip-gram. The numerical data created will be used in the 4 different types of clustering algorithm which is divided into different groups based on the numerical data based on sample data of Netherland’s telecommunication regarding the solution in telecommunication problem. This research is conducted because there has not been any research on comparing clustering algorithm in telecommunication data and there is no research on comparing 4 types of clustering algorithm on text data. So, 4 types of clustering algorithm which are Centroid-based clustering, Density-based Clustering, Hierarchical Clustering, and Distribution-based Clustering used for clustering the text documents. The result then is evaluated by S_Dbw and observation of the result from each algorithm. The result of the research showed that centroid-based has the best performance compared to other clustering algorithms based on the telecommunication data while Hierarchical algorithm is the most consistent on grouping the data.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science BSc (56964)
Link to this item:http://purl.utwente.nl/essays/87033
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page