Comparable estimation of network power for chi-squared Pearson functional networks and Bayes hyperbolic functional networks while processing biometric data

This paper aims at the comparison of the network power for Pearson-Hamming networks built using the chi-squared functional set, and Bayes-Hamming networks built using the hyperbolic functional set. To configure these networks a correlation matrix of biometric data is calculated. At the nest step the data are sorted. Low-correlated data are converted with Pearson-Hamming networks, high-correlated data are converted using Bayes-Hamming networks. The detection of a pair with high-correlated parameters r ≈ 0.99 is equal to the detection of approximately 9 pairs of low-correlated parameters r ≈ 0. The power gain for Pearson-Hamming and Bayes-Hamming networks are comparable. Low-correlated parameters dominate but they are less significant than high-correlated parameters.