PENGARUH REDUKSI FITUR DENGAN METODE ISOMETRIC FEATURE MAPPING PADA PENGKLASTERAN DOKUMEN BERDIMENSI TINGGI

FEBRIYANTO, RAHMAD TIRTA and Jambak, M. Ihsan and Kurniati, Rizki (2019) PENGARUH REDUKSI FITUR DENGAN METODE ISOMETRIC FEATURE MAPPING PADA PENGKLASTERAN DOKUMEN BERDIMENSI TINGGI. Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021181520002.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
Preview	Text RAMA_55201_09021181520002_0205046801_8822870018_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB) | Preview
	Text RAMA_55201_09021181520002_0205046801_8822870018_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (195kB) | Request a copy
	Text RAMA_55201_09021181520002_0205046801_8822870018_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (739kB) | Request a copy
	Text RAMA_55201_09021181520002_0205046801_8822870018_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021181520002_0205046801_8822870018_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (724kB) | Request a copy
	Text RAMA_55201_09021181520002_0205046801_8822870018_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (93kB) | Request a copy
	Text RAMA_55201_09021181520002_0205046801_8822870018_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (105kB) | Request a copy
	Text RAMA_55201_09021181520002_0205046801_8822870018_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (204kB) | Request a copy
	Text RAMA_55201_09021181520002_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (12MB) | Request a copy

Abstract

Implementasi pengklasteran diantaranya adalah pengelompokkan dokumen dengan mengubah dokumen teks menjadi data numerik. Setiap kata yang berbeda dalam dokumen teks akan menjadi atribut baru pada data numerik, sehingga data memiliki banyak dimensi. Kinerja algoritma pengklasteran konvensional tidak efektif dan efisien dalam menangani data berdimensi tinggi. Oleh karena itu, diperlukan teknik ekstraksi fitur menggunakan Principal Component Analysis (PCA) dan Isometric Feature Mapping (ISOMAP) untuk mengurangi dimensi pada data. Penelitian ini membandingkan teknik esktraksi fitur pada pengklasteran k-Means. Dari penelitian ini, didapatkan hasil rata-rata DBI ISOMAP+k-Means sebesar 15,652 dan PCA+k-Means sebesar 17,650 yang tidak ada perbedaan signifikan, tetapi metode ISOMAP+k-Means lebih efisien dengan rata-rata waktu komputasi pengklasteran 0,014 detik dan waktu komputasi reduksi 1,113 detik. . Kata Kunci: Data Berdimensi Tinggi, Ekstraksi Fitur, Principal Component Analysis, Isometric Feature Mapping, k-Means Clustering.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Data Berdimensi Tinggi, Ekstraksi Fitur, Principal Component Analysis, Isometric Feature Mapping, k-Means Clustering.
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation. Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA76.Z55 Apache Hadoop (Computer file) Electronic data processing--Distributed processing. File organization (Computer science) Data mining. Streaming technology (Telecommunications)
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Users 2092 not found.
Date Deposited:	26 Sep 2019 07:22
Last Modified:	26 Sep 2019 07:22
URI:	http://repository.unsri.ac.id/id/eprint/8981

Actions (login required)

View Item