Setyawan, Handi Fajar (2023) ANALISIS PERFORMANSI CONVOLUTION NEURAL NETWORK (CNN) DAN NEURAL NETWORK (NN) TERHADAP IDENTIFIKASI TUJUH JENIS BUAH PIR. Undergraduate thesis, UPN Veteran Jawa Timur.
|
Text (Cover)
17081010012_cover.pdf Download (2MB) | Preview |
|
|
Text (BAB 1)
17081010012_Bab1.pdf Download (927kB) | Preview |
|
![[img]](http://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (BAB 2)
17081010012_Bab2.pdf Restricted to Registered users only until 23 November 2025. Download (3MB) | Request a copy |
|
|
Text (BAB 3)
17081010012_Bab3.pdf Restricted to Registered users only until 23 November 2025. Download (1MB) | Request a copy |
|
|
Text (BAB 4)
17081010012_Bab4.pdf Restricted to Registered users only until 23 November 2025. Download (5MB) | Request a copy |
|
|
Text (DAFTAR PUSTAKA)
17081010012_DaftarPustaka.pdf Download (141kB) | Preview |
|
|
Text (BAB 5)
17081010012_Bab5.pdf Download (362kB) | Preview |
|
|
Text (LAMPIRAN)
17081010012_Lampiran.pdf Restricted to Registered users only until 23 November 2025. Download (4MB) | Request a copy |
Abstract
Pears are one of the fruits that people often consume Indonesia is the Asian pear (Pyrus pyrifolia) because pears have characteristics which is sweet, sour and crunchy is a fruit that is popular in Indonesia, because The high consumption of pears in Indonesian society is proven by there were imports from China, Australia, South Korea and America which reached 69 thousand tons in 2012. Apart from that, this fruit has a distinctive taste and is identical to lots of water, salt and sweetness. There are also nutrients and various kinds of vitamins from fruit that lives in tropical soil, including A, B1, B2, C, E, K, niacin, acid pantothenate, and folacin. The method used is Neural Convolution Network (CNN) and Neural Network (NN) using architecture from both methods to get a comparison of the accuracy level results. The accuracy level parameters are divided into 6 parts for the train data, namely 50%, 55%, 60%, 65%, 70%, and 75% then the results of the training will brings up the level of accuracy of the two algorithms with the level of accuracy highest for CNN (98%) and NN (93%) methods. After doing training The test process will be carried out using a confusion matrix for get validation results, for the CNN method the accuracy level is 97%, precision level 97%, sensitivity_recall 97%, f1_score 97% and for the NN method accuracy level got 86%, precision level 89%, sensitivity_recall 86%, f1_score 86%
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA76.6 Computer Programming Q Science > QA Mathematics > QA76.87 Neural computers T Technology > T Technology (General) |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Handi Fajar Setyawan Handi Fajar Setaywan | ||||||||||||
| Date Deposited: | 23 Nov 2023 05:14 | ||||||||||||
| Last Modified: | 23 Nov 2023 05:14 | ||||||||||||
| URI: | http://repository.upnjatim.ac.id/id/eprint/18795 |
Actions (login required)
 |
View Item |