Malicious odor item identification using an electronic nose based on support vector machine classification

The aim of this study is to develop an electronic nose for identifying the spoiled meat stocked inside a refrigerator. Electronic nose analyses the samples of beef and fish and applies a classifier named support vector machine (SVM) to identify the meat creating malodor. To evaluate, the experiment is performed for a week. The results indicate that SVM classifier exhibits good generalization performance and enable accuracy rate of almost 94.5 % for both beef and fish. This means that SVM is an effective pattern classification technique for spoiled meat identification using electronic nose.