Pengaruh Lingkungan Terhadap Deteksi Dan Klasifikasi Kemurnian Daging Sapi Menggunakan Electronic Nose

Wakhid, Sulaiman (2021) Pengaruh Lingkungan Terhadap Deteksi Dan Klasifikasi Kemurnian Daging Sapi Menggunakan Electronic Nose. Masters thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
09211950053005-Master_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (2MB) | Request a copy

Abstract

Daging sapi merupakan komoditas yang memiliki tingkat konsumsi cukup tinggi diberbagai negara. Beberapa negara memiliki standar dan aturan tersendiri dalam menjamin kualitas daging sapi yang layak konsumsi. Salah satu ukuran kelayakan konsumsi daging adalah dari tingkat kemurnian daging sapi. Namun saat ini masih ada isu manipulasi kemurnian daging sapi dengan mencapurkan daging babi . Hal ini tentu menjadi isu yang merugikan bagi konsumen daging sapi. Isu tersebut dapat di atasi dengan cara pendeteksian daging sapi melalui perangkat electronic nose. Deteksi kemurnian daging sapi dengan electronic nose dapat dilakukan dengan cara mengambil senyawa gas dari aroma daging. Dalam proses pengambilan gas faktor lingkungan seperti konsentrasi gas, suhu, dan kondisi ruangan dapat mempengaruhi keakurasian dan kualitas data yang diambil. Penelitian ini bertujuan untuk mengetahui hasil deteksi dan klasifikasi kemurnian daging sapi melalui electronic nose yang dipengaruhi oleh faktor konsentrasi gas bedasarkan ukuran sample chamber. Sehingga dapat diketahui salah satu pengaruh lingkungan berupa konsentrasi gas terhadap kesiapan electronic nose untuk kondisi penggunaan lingkungan nyata. Pada penelitian ini digunakan sampel campuran daging sapi dan daging babi sebanyak 7 kelas perbandingan untuk klasifikasi. Untuk mengetahui pengaruh lingkungan maka diakukan uji coba yang berfokus pada pengaruh konsentrasi gas berdasarkan ukuran sample chamber terhadap akurasi dari klasifikasi. Adapun metode klasifikasi yang digunakan dalam penelitian ini menggunakan Decision Tree Classifier, Logistic Regression, Support Vector Machine (SVM), dan Artificial neural network (ANN) yang diperkuat dengan metode Ensemble. Berdasarkan hasil penelitian menunjukkan adanya pengaruh lingkungan berupa konsentrasi gas terhadap hasil klasifikasi dan deteksi. Dimana konsentrasi gas yang tinggi pada sample chamber terkecil dapat memperkuat hasil klasifikasi hingga tingkat akurasi 95,71%. =================================================================================================== Beef is a high level of consumption commodity in various countries. Some countries have their standards and rules for ensuring the beef quality that is fit for consumption. One measure of the feasibility of meat consumption is the level of purity of beef. However, currently, there is still the issue of manipulating beef purity by mixing pork and beef. This case is a detrimental issue for beef consumers. Recently this issue can be resolve by detecting beef through an electronic nose device. Detection of the purity of beef with an electronic nose can be done by taking gas compounds from the aroma of the meat. In the taking gas process, environmental factors such as gas concentration, temperature, and room conditions can affect the accuracy and quality of the data taken. This study aims to determine the results of detection and classification of beef purity through an electronic nose which is influenced by the gas concentration factor based on the sample chamber size. So that it can be seen that one of the environmental influences in the form of gas concentration on the readiness of the electronic nose for real environmental use conditions. In this study, a sample of a mixture of beef and pork was used as many as 7 comparison classes for classification. To determine the effect of the environment, a trial was conducted that focused on the effect of gas concentration based on the size of the sample chamber on the accuracy of the classification. The classification method used in this study uses Decision Tree Classifier, Logistic Regression, Support Vector Machine (SVM), and Artificial Neural Network (ANN) which is strengthened by the Ensemble method. Based on the results of the study showed that there was an environmental influence in the form of gas concentration on the classification and detection results. Where the high gas concentration in the smallest sample chamber can strengthen the classification results to an accuracy level of 95.71%.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Daging, Electronic Nose, Klasifikasi, Sensor, Machine Learning, Beef, E-nose, Classification, Sensor, Machine Learning
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning.
Q Science > Q Science (General) > Q337.5 Pattern recognition systems
Q Science > QA Mathematics > QA336 Artificial Intelligence
T Technology > T Technology (General)
T Technology > TP Chemical technology > TP370 Food processing and manufacture
Divisions: Faculty of Creative Design and Digital Business (CREABIZ) > Technology Management > 61101-(S2) Master Thesis
Depositing User: Sulaiman Wakhid
Date Deposited: 13 Aug 2021 15:52
Last Modified: 13 Aug 2021 15:52
URI: https://repository.its.ac.id/id/eprint/86287

Actions (login required)

View Item View Item