ALYA RAMADHANNI, NABILLA (2020) KLASIFIKASI AROMA BUAH DENGAN SENSOR GAS MENGGUNAKAN PENDEKATAN STATISTIKA DAN UJI SIMILARITAS DENGAN METODE K-NEAREST NEIGHBOR (KNN) BERBASIS ARDUINO. Undergraduate thesis, Universitas Stikubank.
![[img]](http://eprints.unisbank.ac.id/style/images/fileicons/application_pdf.png) |
PDF (HLM JUDUL)
Download (1MB) |
|
PDF (ABSTRAK)
Download (30kB) |
|
PDF (BAB I)
Download (74kB) |
|
PDF (BAB II)
Restricted to Repository staff only Download (84kB) |
|
PDF (BAB III)
Restricted to Repository staff only Download (389kB) |
|
PDF (BAB IV)
Restricted to Repository staff only Download (720kB) |
|
PDF (BAB V)
Restricted to Repository staff only Download (7kB) |
|
PDF (DAFTAR PUSTAKA)
Download (6kB) |
|
PDF (LAMPIRAN)
Restricted to Repository staff only Download (996kB) |
Abstract
Hampir semua sektor telah menggunakan otomatisasi mekanis dan digitalisasi teknologi. Banyak industri ingin mengefisiensi sumber daya sehingga dibutuhkannya alat guna membantu jalannya produksi dalam industri, contohnya industri makanan minuman, parfum, obat-obatan dan lain-lain. Penelitian ini akan membuat sebuah alat untuk mengklasifikasi dan menguji similaritas pada aroma buah dengan sensor gas MQ 2, MQ 3, MQ 4, MQ 6, MQ 7, MQ 8 dan arduino uno. Kemudian dari sistem yang telah dibuat diproses mengguankan software Arduino IDE , untuk klasifikasi menggunakan pendekatan statistika dan untuk uji similaritas menggunakan metode k-nearest neighbor (knn) . Data yang digunakan sebanyak 800 data sampel dan 20 data uji, dengan menghitung rata-rata dari nilai tiap sensor dan dicari nilai dengan jarak terdekat (euclidean) sehingga hasil dari penelitian ini menyimpulkan bahwa dari 6 sensor gas MQ menghasilkan pesentase similaritas aroma apel sebesar 60% , presentase similaritas aroma jeruk 80%, presentase similaritas aroma campuran apel dan jeruk sebesar 60% , presentase similaritas aroma strawberry sebesar 100% dan presentase keakurasian sebesar 75%. Abstrak Almost all sectors have used mechanical automation and technological digitisation. Many industries want to efficiency resources so that the need for tools to help the course of production in the industry, for example beverage food industry, perfume, medicines and others. This research will create a tool to classify and test the similarity on the scent of fruit with a gas sensor MQ 2, MQ 3, MQ 4, MQ 6, MQ 7, MQ 8 and Arduino Uno. Then from the system that has been made processed software Arduino IDE, for classification using a statistical approach and to test similarity using the method K-nearest neighbor (KNN). Data used as much as 800 sample data and 20 test data, by calculating the average of the value of each sensor and sought value with the closest distance (Euclidean) so that the results of this study concluded that from 6 MQ gas sensors produce a pesentase Apple Aroma is 60%, percentage of the flavor of citrus Aroma 80%, Percentage of apple and citrus aroma of mixture is 60%, a percentage of the aroma of strawberry with 100% and a percentage of accuracy of 75%.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | NIM:16.01.53.0136 SKR.I.05.01.1856 |
| Uncontrolled Keywords: | Statistical approaches, KNN, classification of fruity aromas. |
| Subjects: | Q Science > QA Mathematics Q Science > QA Mathematics > QA76 Computer software |
| Faculty / Institution: | Fakultas Teknologi Informasi > Program Studi Teknik Informatika |
| Depositing User: | Ani Mariawati |
| Date Deposited: | 08 May 2020 05:44 |
| Last Modified: | 08 May 2020 05:44 |
| URI: | https://eprints.unisbank.ac.id/id/eprint/6720 |
Actions (login required)
 |
View Item |
