Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/21315
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Muhathir | - |
dc.contributor.advisor | Syah, Rahmad | - |
dc.contributor.author | Suprayogo, Rizki | - |
dc.date.accessioned | 2023-10-02T08:17:01Z | - |
dc.date.available | 2023-10-02T08:17:01Z | - |
dc.date.issued | 2023-01-19 | - |
dc.identifier.uri | https://repositori.uma.ac.id/handle/123456789/21315 | - |
dc.description | 61 Halaman | en_US |
dc.description.abstract | Boosting algorithm merupakan metode pembelajaran ensemble yang mengabungkan learner yang lemah menjadi learner yang kuat untuk meminimalkan kesalahan. Setiap metode mencoba mengkompensasi kelemahan dengan setiap iterasi, aturan lemah dari masing-masing classifier digabungkan untuk membentuk satu aturan akurasi yang kuat. Bosting algorithm memiliki 4 jenis metode yaitu Adaptive Boosting (Adaboost), Gradient Boosting, Extreme Gradient Boosting (XGBoost), dan Light Gradien Boosting Machine (LightGBM). Dikarenakan boosting alghorithm memiliki keunikan dari masing-masing jenis metode. Maka penelitian ini akan di uji coba pada kasus penyakit daun mangga yang terdiri dari penyakit daun mangga Capmodium dan penyakit daun mangga Collectricum dengan memanfaatkan bantuan ektraksi fitur Histogram of Oriented Gradient (HOG). Hasil akurasi tertinggi pada klasifikasi penyakit daun mangga dengan memanfaatkan ektraksi fitur algoritma HOG pada XGBoost mencapai 95%, sedangkan Adaboost mencapai 85%, kemudian Gradient Boosting mencapai 89% , selanjutnya LightGBM mencapai 91%. Boosting algorithm is an ensemble learning method that combines weak learners to become strong learners to minimize errors. Each method tries to compensate for weaknesses with each iteration, the weak rules of each classifier combine to form one strong accuracy rule. There are 4 types of boosting algorithms, namely Adaptive Boosting (Adaboost), Gradient Boosting, Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM). Because the boosting alghorithm has the uniqueness of each type of method. So this research will be tested on cases of mango leaf disease consisting of Capmodium mango leaf disease and Collectricum mango leaf disease by utilizing the help of Histogram of Oriented Gradient (HOG) feature extraction. The highest accuracy results in the classification of mango leaf disease by utilizing the HOG feature extraction algorithm on XGBoost reach 95%, while Adaboost reaches 85%, then Gradient Boosting reaches 89% , then LightGBM reaches 91%. | en_US |
dc.language.iso | id | en_US |
dc.publisher | UNIVERSITAS MEDAN AREA | en_US |
dc.relation.ispartofseries | NPM;178160118 | - |
dc.subject | penyakit daun mangga | en_US |
dc.subject | boosting algorithm | en_US |
dc.subject | hog | en_US |
dc.subject | mango leaf disease | en_US |
dc.title | Analisis Boosting Algorithm dan Hog dalam Klasifikasi pada Penyakit Daun Mangga | en_US |
dc.title.alternative | Analysis of Boosting Algorithm and Hog in Classification of Mango Leaf Diseases | en_US |
dc.type | Skripsi Sarjana | en_US |
Appears in Collections: | SP - Informatic Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
178160118 - Rizki Suprayogo - Fulltext.pdf | Cover, Abstract, Chapter I, II, III, V, Bibliography | 1.56 MB | Adobe PDF | View/Open |
178160118 - Rizki Suprayogo - Chapter IV.pdf Restricted Access | Chapter IV | 478.31 kB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.