Insulators' flashover due to pollution is the biggest challenge of power systems. This phenomenon will become more intense with the aging of the insulator. Precise and prompt determination of the severity of pollution and the insulator status in terms of aging is essential for preventing pollution flashovers. Therefor monitoring insulator leakage current is essential for preventing the pollution flashover phenomenon. This article proposes innovative criteria based on the insulator's leakage current analysis for the classification of aging and pollution severity. The leakage current signal of the virgin, fully aged, and asymmetric aged insulators have been examined under different pollution conditions. Based on the experimental results analysis, two new criteria have been proposed for assessing the silicon rubber insulator condition which have resulted from the first-order harmonic of leakage current (It1), the third and fifth-order harmonics ratio of leakage current (It3/5), and the cosine of first-order harmonic phase angle (PI). Analyzing the proposed criteria demonstrates that two proposed criteria have a strong correlation to aging and pollution, respectively. These two new criteria can be implemented to identify the early signs of pollution and aging of insulators. The ability of proposed criteria has been established based on random forest, support vector machine, decision tree algorithm, multi-layer perceptron, and decision-making level fusion in Python language. The obtained criteria based on experimental data in different conditions have been employed for classifier training which can be applied in industrial settings without the need for laborious data labeling, ensuring fast and accurate outcomes.