Sistem Cerdas Pendeteksi Zat Pengawet Berbahaya Pada Susu Sapi Konsumsi Dengan Electronic Nose Menggunakan Machine Learning

Ramadhani, Donny Kurnia (2022) Sistem Cerdas Pendeteksi Zat Pengawet Berbahaya Pada Susu Sapi Konsumsi Dengan Electronic Nose Menggunakan Machine Learning. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05311840000004-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Susu sapi memiliki banyak manfaat bagi konsumsi manusia, seperti membantu pertumbuhan tulang, menyehatkan jantung, menetralisir racun, dan bahkan untuk mengurangi risiko kanker. Umumnya, susu hanya mengandung sebagian kecil formalin yang terbentuk secara alami di tubuh binatang (dalam kasus ini, sapi). Namun, masa simpan susu sapi murni cukup singkat yang memberikan insentif kepada oknum-oknum tidak bertanggung jawab. Pencampuran formalin ke susu sapi dapat mengakibatkan dampak yang negatif apabila dikonsumsi oleh manusia. Deteksi susu yang tercampur formalin dapat dilakukan dengan mencium aroma susu, namun hal ini sangat berbahaya dilakukan karena mencium bau formalin secara langsung dapat mengakibatkan efek samping yang merusak tubuh manusia. Maka dari itu, penulis merancang sebuah sistem cerdas pendeteksi zat pengawet berbahaya berbasis alat electronic nose yang dilengkapi dengan 6 sensor gas (3 sensor TGS dan 3 sensor MQ) yang dilatih menggunakan machine learning yang dapat digunakan untuk mengidentifikasi kandungan formalin dalam susu sapi konsumsi. Untuk mempermudah penelitian di masa mendatang, penulis juga menguji pengaruh masing-masing sensor gas yang digunakan untuk mendapatkan pengaruh sensor sehingga apabila terdapat sensor yang kurang berpengaruh dapat dihilangkan untuk menghemat biaya. Hasil penelitian berupa hasil perbandigan 3 model dengan algoritma Support Vector Machine (SVM), Gaussian Naïve Bayes (GNB), dan K-Nearest Neighbors (KNN). Menurut hasil perbandingan, model dengan tingkat akurasi tertinggi memiliki nilai akurasi mencapai 97,21% ketika melakukan deteksi terhadap formalin menggunakan alat electronic nose. Sensor MQ138 adalah sensor dengan pengaruh terbesar terhadap pengambilan keputusan model klasifikasi.
====================================================================================================================================
Cow's milk has many benefits for human consumption, such as helping bone growth, nourishing the heart, neutralizing toxins, and even to reduce the risk of cancer. Generally, milk contains only a small amount of formaldehyde that is naturally occurring in the animal's body (in this case, cows). However, the shelf life of pure cow's milk is quite short which gives incentive to irresponsible people. The mixing of formaldehyde into cow's milk can have negative effects if consumed by humans. Detection of milk mixed with formalin can be done by smelling the aroma of milk, but this is very dangerous to do because smelling formalin directly can cause side effects that damage the human body. Therefore, the authors designed an intelligent system for detecting harmful preservatives based on an electronic nose device equipped with 6 gas sensors (3 TGS sensors and 3 MQ sensors) trained using machine learning that can be used to identify formaldehyde content in consumed cow's milk. To facilitate future research, the author also tests the effect of each gas sensor used to get the effect of the sensor so that if there are sensors that are less influential, they can be removed to save costs. The research results are in the form of comparison results of 3 models with Support Vector Machine (SVM), Gaussian Naïve Bayes (GNB), and K-Nearest Neighbors (KNN) algorithms. According to the comparison results, the model with the highest accuracy rate has an accuracy value of 97.21% when detecting formaldehyde using an electronic nose device. The MQ138 sensor is the sensor with the greatest influence on the classification model's decision making.

Item Type:	Thesis (Other)
Additional Information:	RSTI 006.3 Ram s-1 2022
Uncontrolled Keywords:	Susu Sapi, Electronic Nose, Formalin, Machine Learning, Cow’s Milk, Electronic Nose, Formalin, Machine Learning
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information Technology > 59201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	22 Apr 2026 04:15
Last Modified:	22 Apr 2026 04:15
URI:	http://repository.its.ac.id/id/eprint/132855

Actions (login required)

View Item