The phenomenon of physiological change in plants hyperpolarizes membrane potentials showing a significant change in electrophysiology. For plants to survive, develop, and reproduce, the proper ratio of nutrients is required. Malnourished plants exhibit unfavorable changes in their physiology. Nutrient deficiencies are one of the primary abiotic stresses occurred in plants which are commonly identified and classified through various hyperspectral imaging, and pattern recognition methods. In contrast to image-based methods, in this paper, the electrophysiological signal data of tomato plants in multiple nutrient deficiency states are classified. These electrical signals of tomato plants are acquired in healthy and in deficient states of calcium (Ca), manganese (Mn), nitrogen (N), and iron (Fe). The features are extracted by 7-level orthogonal wavelet decomposition using four different wavelets Daubechies (Db), Coiflet (Cf), Symlet (Sym), and Discrete Meyer (Dm). A novel range-based sorted sample clustering method is proposed to boost classification performance. Class assignment is performed in two modes homogeneous and non-homogeneous, where multi-class classification in non-homogeneous class assignment attained 99.4 % accuracy and in homogeneous class assignment a highest accuracy of 96.8 % is attained using the KNN classifier. Identifying nutrient deficiencies through electrophysiological signal analysis dominated the detection through imaging techniques, impacting a drastic change in crop protection practices.