Abdullah Mu'adz Muflih, . (2024) Rancang Bangun E-Nose dengan Variasi Sensor Gas MQ untuk Klasifikasi Kopi. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.
![[img]](http://repository.unj.ac.id/style/images/fileicons/text.png) |
Text
1_Cover.pdf Download (1MB) |
|
Text
2_Bab 1.pdf Download (400kB) |
|
Text
3_Bab 2.pdf Restricted to Registered users only Download (641kB) |
|
Text
4_Bab 3.pdf Restricted to Registered users only Download (865kB) |
|
Text
5_Bab 4.pdf Restricted to Registered users only Download (605kB) |
|
Text
6_Bab 5.pdf Restricted to Registered users only Download (323kB) |
|
Text
7_Daftar Pustaka.pdf Download (406kB) |
|
Text
8_Lampiran.pdf Restricted to Registered users only Download (466kB) |
Abstract
Penelitian ini bertujuan untuk mengembangkan perangkat Electronic Nose (E-Nose) berbiaya rendah berbasis larik sensor gas MQ untuk mengklasifikasikan bubuk kopi Arabika dan Robusta. Data respons sensor terhadap aroma sampel kopi diperoleh menggunakan instrumen E-Nose yang dirancang. Dua metode klasifikasi digunakan yaitu Support Vector Machine (SVM) dan Artificial Neural Network (ANN) dengan ekstraksi fitur nilai maksimum, rata-rata, dan Area Under Curve (AUC). Hasil menunjukkan bahwa model SVM yang dilatih menggunakan metode Grid Search dengan semua fitur menghasilkan akurasi pengujian 100% sedangkan hasil pelatihan dengan 5 fitur terpilih tetap menghasilkan akurasi pengujian yang tinggi sebesar 90%. Tidak berbeda jauh dengan hasil model SVM, model ANN yang dilatih dengan optimasi Grid Search mampu mencapai akurasi 100% untuk semua fitur sedangkan hasil pelatihan dengan 5 fitur terpilih menghasilkan akurasi yang masih tinggi yaitu 90%. Penelitian ini berhasil mengembangkan E-Nose murah yang efisien dalam membedakan aroma kopi dengan akurasi tinggi menggunakan metode pembelajaran mesin. ***** This study aims to develop a low-cost Electronic Nose (E-Nose) device based on MQ gas sensor array to classify Arabica and Robusta coffee grounds. Sensor response data to the aroma of coffee samples were obtained using the designed E-Nose instrument. Two classification methods were used, namely Support Vector Machine (SVM) and Artificial Neural Network (ANN) with maximum, average, and Area Under Curve (AUC) feature extraction. The results showed that the SVM model trained using the Grid Search method with all features produced 100% testing Accuracy while the training results with 5 selected features still produced a high testing Accuracy of 90%. Not much different from the results of the SVM model, the ANN model trained with Grid Search optimization was able to achieve 100% Accuracy for all features while the results of training with 5 selected features produced a still high Accuracy of 90%. This research successfully developed a low-cost E-Nose that is efficient in distinguishing coffee aroma with high Accuracy using machine learning methods.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Additional Information: | 1). Dr.rer.nat. Bambang Heru Iswanto, M.Si. 2). Haris Suhendar, M.Si. |
| Subjects: | Sains > Fisika |
| Divisions: | FMIPA > S1 Fisika |
| Depositing User: | Users 24987 not found. |
| Date Deposited: | 12 Aug 2024 04:52 |
| Last Modified: | 12 Aug 2024 04:52 |
| URI: | http://repository.unj.ac.id/id/eprint/49782 |
Actions (login required)
 |
View Item |
