A tool developed in Matlab for multiple correspondence analysis of fuzzy coded data sets: application to morphometric...

Area of Science:

• Multivariate statistics
• Bioinformatics
• Anthropology

Background:

• Multiple Correspondence Factorial Analysis (MCA) is a multivariate exploratory method for contingency tables.
• Existing software for fuzzy MCA is limited, lacking interactivity and fuzzy windowing capabilities.
• Fuzzy data analysis extends MCA to data with linguistic properties.

Purpose of the Study:

• To present an interactive Matlab software for computing and visualizing fuzzy MCA results.
• To enable users to define and represent fuzzy windowing with pre-defined membership functions.
• To apply the software to analyze morphometric data and compare it with Principal Component Analysis (PCA).

Main Methods:

• Development of a Matlab-based software tool for fuzzy MCA.
• Utilizing pre-defined membership functions based on data distribution histograms.
• Application to a dataset of 150 male Egyptian skulls across five civilization periods.

Main Results:

• The software facilitates the computation and representation of fuzzy MCA results.
• Application to skull morphometrics demonstrated a rapid description of morphological evolution over time.
• Linguistic descriptions derived from fuzzy MCA provided clearer insights than PCA.

Conclusions:

• The developed software enhances the analysis of fuzzy coded data using MCA.
• It offers a more intuitive and rapid method for describing evolutionary trends in morphometric data compared to PCA.
• The tool supports interactive exploration and visualization of fuzzy windowing in multivariate analysis.