Data Mining Techniques in Food Safety

Abstract

Food safety is an important consideration because the reduced quality of freshness may result in food poisoning that can threaten our health. There are many methods used to test the freshness of food such as visual appearance as well using a variety of devices. One of the devices that can be used to test the freshness of food is the E-Nose. E-nose is an instrument that enables the discrimination of gas and odor in food industry for quality and safety purposes. It is a well-established instrument to detect odor and aroma not only in the food industry, but also in health-diagnosis, defense, and environmental industry. Generally, E-nose mimics human olfactory sense to detect and discriminate gasses or volatile organic compound from a few objects such as food, chemicals, explosive etc. Thus, E-nose can be used to measure gas emitted from food due to its ability to measure gas and odor. Principally, the E-nose operates by using a number of sensors to response to molecules from vaporous compound. Each sensor will respond to their specific gas respectively. These sensors are the major component in electronic nose to sense and obtain percentage of gases release by the compound samples. All gases detected by sensors will be recorded, that to be analyzed using classification method. Classification is a way to distinguish a mixture odor/aroma obtained from gas sensors using a method of machine learning. In this paper, we discussed briefly about electronic nose, it’s principle of work and classification method and in order to classify food freshness.