Optimization of Green Tea Qualification Model Based on Feature Selection Using K-Nearest Neighbor Method with Electronic Nose
Tea is a beverage made from tea leaf shoots that are widely consumed by society. In addition, tea is one of the ex-port goods that generate significant returns for the nation's for-eign exchange. The most popular variety of tea among consum-ers is green tea, which is not only reviving but also...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/282288/1/Yessi%20Ideaningrum%20et%20al-2022-Optimization_of_Green_Tea_Qualification_Model_Based_on_Feature_Selection_Using_K-Nearest_Neighbor_Method_with_Electronic_Nose.pdf |
| _version_ | 1797037571384016896 |
|---|---|
| author | Tanu Wijaya, Yessi Idianingrum Humaira, Shafura Sumanto, Budi |
| author_facet | Tanu Wijaya, Yessi Idianingrum Humaira, Shafura Sumanto, Budi |
| author_sort | Tanu Wijaya, Yessi Idianingrum |
| collection | UGM |
| description | Tea is a beverage made from tea leaf shoots that are widely consumed by society. In addition, tea is one of the ex-port goods that generate significant returns for the nation's for-eign exchange. The most popular variety of tea among consum-ers is green tea, which is not only reviving but also good for the body's health. Gunpowder, Pecco, and Dried are three levels of qualities of green tea that are formed depending on how much tea leaf is extracted. The classification of the quality green tea requires using an objective tool, such as the Electronic Nose, as human tests on aroma did not yield significant results. Sensor readings generate a significant amount of data, which might hin-der the performance of the classification model throughout the data processing and result in a lack of efficacy. This study aims to improve the performance of the classification model by selecting the best combination of features from statistical values. The results show that the combination of two feature configurations, namely kurtosis and standard deviation, provides the best con-tribution with an accuracy of 94.44, precision of 95.24, and sensitivity of 94.44. © 2022 IEEE. |
| first_indexed | 2024-03-14T00:05:20Z |
| format | Conference or Workshop Item |
| id | oai:generic.eprints.org:282288 |
| institution | Universiti Gadjah Mada |
| language | English |
| last_indexed | 2024-03-14T00:05:20Z |
| publishDate | 2022 |
| record_format | dspace |
| spelling | oai:generic.eprints.org:2822882023-11-24T03:15:48Z https://repository.ugm.ac.id/282288/ Optimization of Green Tea Qualification Model Based on Feature Selection Using K-Nearest Neighbor Method with Electronic Nose Tanu Wijaya, Yessi Idianingrum Humaira, Shafura Sumanto, Budi Electrical and Electronic Engineering not elsewhere classified Tea is a beverage made from tea leaf shoots that are widely consumed by society. In addition, tea is one of the ex-port goods that generate significant returns for the nation's for-eign exchange. The most popular variety of tea among consum-ers is green tea, which is not only reviving but also good for the body's health. Gunpowder, Pecco, and Dried are three levels of qualities of green tea that are formed depending on how much tea leaf is extracted. The classification of the quality green tea requires using an objective tool, such as the Electronic Nose, as human tests on aroma did not yield significant results. Sensor readings generate a significant amount of data, which might hin-der the performance of the classification model throughout the data processing and result in a lack of efficacy. This study aims to improve the performance of the classification model by selecting the best combination of features from statistical values. The results show that the combination of two feature configurations, namely kurtosis and standard deviation, provides the best con-tribution with an accuracy of 94.44, precision of 95.24, and sensitivity of 94.44. © 2022 IEEE. 2022 Conference or Workshop Item PeerReviewed application/pdf en https://repository.ugm.ac.id/282288/1/Yessi%20Ideaningrum%20et%20al-2022-Optimization_of_Green_Tea_Qualification_Model_Based_on_Feature_Selection_Using_K-Nearest_Neighbor_Method_with_Electronic_Nose.pdf Tanu Wijaya, Yessi Idianingrum and Humaira, Shafura and Sumanto, Budi (2022) Optimization of Green Tea Qualification Model Based on Feature Selection Using K-Nearest Neighbor Method with Electronic Nose. In: 8th International Conference on Science and Technology (ICST). https://ieeexplore.ieee.org/document/10136265 |
| spellingShingle | Electrical and Electronic Engineering not elsewhere classified Tanu Wijaya, Yessi Idianingrum Humaira, Shafura Sumanto, Budi Optimization of Green Tea Qualification Model Based on Feature Selection Using K-Nearest Neighbor Method with Electronic Nose |
| title | Optimization of Green Tea Qualification Model Based on Feature Selection Using K-Nearest Neighbor Method with Electronic Nose |
| title_full | Optimization of Green Tea Qualification Model Based on Feature Selection Using K-Nearest Neighbor Method with Electronic Nose |
| title_fullStr | Optimization of Green Tea Qualification Model Based on Feature Selection Using K-Nearest Neighbor Method with Electronic Nose |
| title_full_unstemmed | Optimization of Green Tea Qualification Model Based on Feature Selection Using K-Nearest Neighbor Method with Electronic Nose |
| title_short | Optimization of Green Tea Qualification Model Based on Feature Selection Using K-Nearest Neighbor Method with Electronic Nose |
| title_sort | optimization of green tea qualification model based on feature selection using k nearest neighbor method with electronic nose |
| topic | Electrical and Electronic Engineering not elsewhere classified |
| url | https://repository.ugm.ac.id/282288/1/Yessi%20Ideaningrum%20et%20al-2022-Optimization_of_Green_Tea_Qualification_Model_Based_on_Feature_Selection_Using_K-Nearest_Neighbor_Method_with_Electronic_Nose.pdf |
| work_keys_str_mv | AT tanuwijayayessiidianingrum optimizationofgreenteaqualificationmodelbasedonfeatureselectionusingknearestneighbormethodwithelectronicnose AT humairashafura optimizationofgreenteaqualificationmodelbasedonfeatureselectionusingknearestneighbormethodwithelectronicnose AT sumantobudi optimizationofgreenteaqualificationmodelbasedonfeatureselectionusingknearestneighbormethodwithelectronicnose |