Team 28

The demand of Electronic Tongue started to appear in the late 1980s and 1990s. From this time the investigation rate had increased in a noticeable way. Also, more work has been published through the years (1996-2020). Even though certain review papers have been published, the majority of them have been focused on specialized domains of application, such as foods or the pharmaceutical business. The very first testing papers were done on primary liquids such as pure water, tap water, milk, and fruit juice such as the strawberry juice. Other assessments place a greater emphasis on specific types of sensors (voltammetric and potentiometric ETS). As a result, the purpose of this work is to fill in the gaps and to provide the most recent developments in various ET systems. Our system is used to classify different types of liquids such as tea, coffee and milk using electrodes as sensors.

Beverage quality and freshness monitoring are of importance to both consumers and the beverage industry. Quality has become a big issue for consumers, especially since the corona pandemic, as their comprehension of all aspects of liquid quality has grown. The majority of liquid identification and quality control approaches take time and necessitate the use of sophisticated equipment and trained staff. Alternate analytical methods should be supplied as a result of these restrictions, and the use of the electronic tongue is a fantastic example of such an alternative strategy. Otto and Thomas presented the first multi-sensor array-based liquid analysis device in 1985. since that time There have been a few gadgets of this type on the market, called “electronic tongues.” The term ‘electronic tongue’ is the most popular in European articles, followed by ‘taste chip’ in American papers and ‘taste sensor/taste system’ in Asian journals. A “taste sensor” is a device that uses technology to classify basic taste sensations and compares the results to those of a human panel. The integration of information from numerous disciplines of research, including sensory technologies, pattern recognition methods, and artificial intelligence, is required for the development of such systems. The (LVQ) neural network, weighted k-nearest neighbours, and Mahalanobis distance were designed to compare the results of the LVQ neural network classifier. The LVQ neural network classifier outperformed the competition with a classification rate of 97.9 %

Mimicking human taste has a huge advantage when it comes to poisonous or repetitive tasting. We can’t risk testing food quality on humans or even animals. The problem is that the food may be rotten or spoiled which may cause the taster huge side effects that may affect his health as well. So, having a system that tests without affecting anyone is such a great thing. In some conditions we can’t rely on human taste as in case of automatic process control especially on an industrial scale and also in economic reasons, defined in terms of time or financial expenses.