Tea is the most popular hot beverageworldwide. In 2020, the value of the global tea market was almost USD 200 billion, and is estimated to reach up to USD 318 billion by the year 2025. Tea has been included as part ofa regular diet for centuries because of its various health benefits. However, tea is acidic, and over-consumption causes heat problems, disturbance of the sleep cycle, tooth erosion, and low calcium absorption in the body. Strong tea concentration is very harmful and toxic. The safe consumption of tea should be guaranteed. The treatment applied in this research work is on sensory mechanisms and Arduino UNO. The objective of this paper is to find out community interest in a particular tea species and inform them about tea overdose.The acidity is mapped with tea taste in terms of strong, medium, and low flavors. Based on the data analysis, the results differentiatethe acidity level of black tea (pH: 3.89–4.08) as very high, green tea (pH: 4.68–4.70) is in the 2nd position, and the energy drink Herbalife Nutrition (pH: 5.59–5.64) is the least acidic comparatively, with a proportion ratio 1:10 of tea to water. Experimental analysis reveals that in the additives, lemon is most acidic, followed byginger, lemongrass, and Tulasi.

India is the second-largest tea producer and consumer in the world after China. In 2017, the Indian tea market size accounted for 130 billion Indian rupees. An estimated global tea market size was at USD 13.31 billion in 2019, and the expected compound annual growth rate is 5.5% up to the year 2025. India can grab worth tea market size globally by making market strategies with AI and ML-based demonstrations for the unique identity of tea flavor. Conventional instruments available are not handy, time-consuming and require a skilled person to operate. The tea attributes should be digitally recognizable before purchase from the consumer's perspective, significantly enlarging the tea market circle. In the paper, the comprehensive review about an artificial perception of tea has been briefly discussed. Three major attributes of the tea sample, its taste, smell, and color, are under consideration. With the help of various sensors, the attributes of liquefied tea samples had got converted into their digital signature. By analyzing the correlation of them with the pattern recognition, their classification had done. The electronic feature fusion of tea liquor attributes may cause handling issues with the formation of redundant data. So this paper explains the method and guidelines of an application of specific filters which remove the redundant data. The constructive sample data can establish the decision matrix for correlation. With the established decision matrix, précised test prediction can be achieved for the tea sample based on correlation and regression. The limitations and glitches of the conventional instruments for an artificial perception have been discussed in-depth for possible improvement. The paper ends with a bibliometric analysis of the topic "artificial taste perception of tea," which had derived from the standard repository of Web of Science. The bibliometric analysis is very useful to showcase the current research trends in the artificial taste perception of tea.

No abstract