Performance Evaluation of a Novel iTongue for Indian Black Tea Discrimination

In this work, the multi-electrode-single-frequency (MESF), multi-frequency-single-electrode (MFSE), and multi-frequency-multi-electrode (MFME) impedance responses of an impedance-Tongue reported previously, are evaluated for their discriminability of Indian Black Teas. Principal component analysis (PCA) in conjunction with a cluster validity measure, Davis-Bouldin Index (DBI), has been used for discriminability evaluation. The discriminabilities of electrode specific frequency segments chosen by optimization algorithms, namely, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) have also been evaluated. The results show that the MFSE impedance response of Gold electrode gives the best discriminability without compromising the system complexity as against MESF, MFSE, MFME, and GA/PSO-optimized response. The results also suggest that the cross-sensitivity of electrodes may be enhanced by selecting optimum frequencies and/or electrodes, paralleling the practice of modifying the electrodes. This opens up a new approach towards qualitative and quantitative analysis of complex liquids.