Tazi, Imam and Suyono, Suyono (2016) Otentikasi dan klasifikasi berbagai minyak nabati dan hewani berbasis penginderaan hidung elektronik menggunakan metode linear discriminant analysis (LDA). Research Report. Lembaga Penelitian dan Pengabdian kepada Masyarakat UIN Maulana Malik Ibrahim, Malang. (Unpublished)
|
Text
imamtazi-rpiinter-2016.pdf Download (1MB) | Preview |
Abstract
The testing system based on the electronic nose has 10 sensor array that was used for pattern classification smell of various cooking oils. The sample were a fortune-branded cooking oil palm, palm oil brands horse, palm oil rose brand, oil olive oil, canola flower oil, lard. The sensor array was formed of 10 gas sensors MQ and TGS. Testing was be done by taking a random sample of cooking oil from traditional markets and modern markets. The sensor response was evaluated using a method of pattern recognition Linear Discriminant Analysis (LDA).
The LDA data processing was using stepwise method for selecting discriminator variables that have contributed in the prediction model discriminant function. Based on the first discriminant function have a value of 94.8%, which means a variant to the first discriminant function was sufficient to classify and distinguish the smell of all six samples of cooking oil. The second discriminant function was showed the variance 5.2% which was less able to distinguish the smell of cooking oil data conditions. In general, the equipment was able to detect the scent of various cooking oils. For the classification of lard and olive oil, the equipment can distinguish the scent of cooking oil than palm oil and canola flowers.
| Item Type: | Research (Research Report) |
|---|---|
| Subjects: | 10 TECHNOLOGY > 1003 Industrial Biotechnology > 100304 Industrial Biotechnology Diagnostics (incl. Biosensors) |
| Divisions: | Faculty of Mathematics and Sciences > Department of Physics |
| Depositing User: | Miftahus Sholehudin |
| Date Deposited: | 10 Jan 2017 13:07 |
Downloads
Downloads per month over past year
Origin of downloads
Actions (login required)
 |
View Item |
Altmetric
Altmetric
