Sensitivity analysis for correspondence analysis and visualization

Correspondence analysis is frequently utilized in data and text mining. It will deepen the global understanding on the characteristics of accumulated text information and may lead to new knowledge discovery. We have so far focused on the text information of higher education institution such as syllabus. We have conducted research on collecting and analysing such information, and also pursuing research on text mining techniques to analyze and visualize the information for grasping the global characteristics of curricula of higher education. However, in case of data variation, result of correspondence analysis might be greatly changed. This article presents mathematical consideration on sensitivity of correspondence analysis and visualization method, and an application to comparative analysis of curricula of higher education institutions.