Show simple item record

dc.contributor.advisorSitumorang, Zakarias
dc.contributor.advisorMawengkang, Herman
dc.contributor.authorFachrurrazi, Sayed
dc.date.accessioned2021-08-04T07:18:31Z
dc.date.available2021-08-04T07:18:31Z
dc.date.issued2011
dc.identifier.urihttp://repositori.usu.ac.id/handle/123456789/39194
dc.description.abstractThis Paper present the analysis of the performance implementing of support vector machine with 11 independent variable and 1 dependent variable. The SVM method with training data (75%) and testing data (25%) able used for classification data domestic flight and internasional flight can be find the best hyperplane rule for 2 classifier. The output for 4 support vector could to find a function to differentiate data classes. which used Structure Risk Minimization (SRM) to find the best hyperplane function to separate two data classes. This research analyzes SVM performance for there aeroplane classifying based on domestic and international rute Some precautions have to be performed in onder to get a good performance i.e preprocessing, kernel application, the appropriate parameter in SVM and feature selection. The study shows that SVM method can be applied to classify IBM. The model accuracy is 84,31% observed using Receiver Operating Characteristic (ROC).en_US
dc.description.abstractTulisan ini menyajikan analisis performansi Support Vector Machine (SVM) dengan 11 variabel bebas dan 1 variabel terikat. Metode SVM dengan data training (75%) dan data testing (25%) yang digunakan pada pengklasifikasian data penerbangan domestik dan data penerbangan internasional untuk menemukan hyperplane terbaik yang memisahkan dua buah kelas. Hasilnya terdapat 4 support vector memberikan informasi yang dibutuhkan untuk menyakinkan bahwa metode SVM bisa sebagai classifier dan dapat memprediksi keakuratan model dengan menggunakan kurva Receiver Operating Characteristic (ROC) untuk melihat akurasi model terbaik. mencapai 84,31%.en_US
dc.language.isoiden_US
dc.publisherUniversitas Sumatera Utaraen_US
dc.subjectklasifikasien_US
dc.subjectmetode support vector machine (SVM)en_US
dc.subjectreceiver operating characteristic (ROC)en_US
dc.subjectsupport vectoren_US
dc.subjecthyperplane terbaiken_US
dc.titlePenggunaan Metode Support Vector Machine (SVM) untuk Mengklasifikasi dan Memprediksi Angkutan Udara Jenis Penerbangan Domestik dan Penerbangan Internasional di Banda Acehen_US
dc.typeThesisen_US
dc.identifier.nimNIM097038014
dc.description.pages58 Halamanen_US
dc.description.typeTesis Magisteren_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record