An Application of Fuzzy Coding in Multiple Correspondence Analysis for Transforming Data from Continuous to Categorical
Zerrin Asan and Sevil Sentürk
In some research, it can be necessary to know the categorical data groups to which the data measured by interval and ratio scale belong. The fuzzy sets principle can be used to transform the original continuous data into categories. Fuzzy coding system is used for the transforming. Then, Multiple Correspondence Analysis (MCA) is proposed to investigate a table where the data are obtained from fuzzy coding. This process is applied for the data collected in many disciplines like biology, ecology, and so on. The aim of this study is to research how the integration of fuzzy coding system and MCA is applied to continuous meteorological data. The triangular membership function is used for transforming the data with this purpose in this application. According to the results, fuzzy coding is more preferable for transforming continuous meteorological data into categories because fuzzy coding explains nonlinear relationships better than linear relationships.
Keywords: Fuzzy coding, multiple correspondence analysis, meteorological categorical data.