PENERAPAN ALGORITMA K-NEAREST NEIGHBOR PADA E-NOSE UNTUK IDENTIFIKASI TAHU BERFORMALIN; IMPLEMENTATION OF K-NEAREST NEIGHBOR ON E-NOSE FOR FORMALINE TOFU IDENTIFICATION

Utami, RAHMA FAUZIA, Danang Lelono

2016 | Skripsi | FMIPA

Tofu is kind of food that contain water and soya bean as a composition, so it will be decomposed in one day. Mixing tofu and formalin is a method to endure the phisycs of tofu, but this method is very dengeraus for human health. Pure tofu and formaline tofu can be identified by colour, for testing, it will be involving eye sensory to see physical characteristics of tofu. Qualitative test with chromothopat acid reagent and quantitative test with spectrophotometer test. All of them involving human tester. However, the tester is subjective. Beside that standard chemical analytic method requires a high cost and need the expert to analyzing. Basically aroma of tofu is determined by volatile. Based on this aroma there are the identification of formalin tofu and pure tofu using an electronic nose (e-nose) by applying the k-Nearest Neighbor method for data analysis. Electronic nose based on unselected gas sensor array has the ability to analyze samples with complex compositions that can be known characteristics and qualitative analysis of the samples. Stimulus aroma is transformed by electronic nose into fingerprint data then it is used by feature extraction process using the integral method. The results of feature extraction is used as training data and tast data for test using k-neasrest neighbor (k-NN) Based on test results with variation of k = 3,5,7,9,11 and 13 , out of pure and formalin tofu 0.5 % - 7 % is obtained accuracy of the test results 96.42 % with training data 56. Then, by increasing the training data to 112 , the tofu identification result to be 100 % . The value k = 3 has been able to present the results of the test well . So that the identification of a pure and formalin tofu with electronic nose using k - NN method has been successfully performed .

Kata Kunci : electronic nose, k-nearest neighbor, tofu, identification