Visualization for University Brand Image Clustering: Comparison between Male and Female Students
Yukari Shirota, Setsuko Katayama, Takako Hashimoto, Basabi Chakraborty · International Journal of Institutional Research and Management · 2017
In this paper, visualization of the results of clustering the brand images of different universities has been presented.Generally high school students care a lot about the brand images of the universities.The brand image is an important criterion for their selection of a particular university to continue their studies.The brand images of universities have been analyzed here by using Bayesian inference.Bayesian inference is widely used in various application fields such as topic extraction.An analysis by topic model has been conducted so far in topic extraction.However, as far as we know, no research that applies the topic model for the clustering of university brand images is available.This is the first application of the topic model in clustering brand images of universities.In this paper, we present the simple topic model visualization tool that we have developed and, as an example, we have applied our developed tool to visualize the clustering of university brand image.When high school students select a university, their university choices greatly depend on the universities' brand images.So the university public relations section needs the survey of the brand images.The clustering results are helpful for the public relations section of the universities to fix up their publicity strategy.The clustering results show that there is a class of most popular and most difficult universities for high school students and there is another class of public universities which can be characterized by high level education and low cost tuition.The outcome of the clustering is quite consistent with our expectation.In this paper comparison of university choices between male and female students is also presented.