A Study on Clustering Performance of Twitter Usage Statistics of State University Rectors in Turkey

Abstract

The purpose of clustering analysis is to group several ungrouped data homogeneously based on their similarities. Clustering analysis is a multivariate statistical method, widely used especially in a variety of disciplines such as medicine, engineering, agriculture and social sciences. This study aims to classify the rectors of state universities in Turkey according to their Twitter use in the first six months of the COVID-19 outbreak. Since COVID-19 cases were first seen in Turkey on 11 March, the data collection interval was set as 11th March – 11th September. Firstly, the rectors were clustered using the Weka program based on Twitter usage data. As a result of the clustering analysis, these rectors were collected in six separate clusters. Then, the performance analysis for the clusters was carried out with the help of the MULTIMOORA method, which is one of the Multi-Criteria Decision-Making approaches. Finally, the Twitter usage performances of the groups were discussed based on the findings from the performance analysis.

PDF
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright (c) 2021 The copyright in this website and the material on this website (including without limitation the text, computer code, artwork, photographs, images, music, audio material, video material and audio-visual material on this website) is owned by the European Journal of Digital Economy Research (EJDER) and its licensors.