PERBANDINGAN ALGORITMA CLUSTERING K-MEANS DAN K-MEDOID UNTUK PERINGKASAN TEKS OTOMATIS ARTIKEL BERITA
| Main Author: | Linarko, Rizqy |
|---|---|
| Format: | Thesis NonPeerReviewed Book |
| Bahasa: | eng |
| Terbitan: |
, 2019
|
| Subjects: | |
| Online Access: |
http://eprints.umm.ac.id/54798/43/PENDAHULUAN.pdf http://eprints.umm.ac.id/54798/1/BAB%20I.pdf http://eprints.umm.ac.id/54798/3/BAB%20II.pdf http://eprints.umm.ac.id/54798/2/BAB%20III.pdf http://eprints.umm.ac.id/54798/4/BAB%20IV.pdf http://eprints.umm.ac.id/54798/7/BAB%20V.pdf http://eprints.umm.ac.id/54798/6/Lampiran.pdf http://eprints.umm.ac.id/54798/44/Poster%20Skripsi.pdf http://eprints.umm.ac.id/54798/ |
Daftar Isi:
- Technological progress is now increasingly rapid, this is marked by increasingly easy access to get information from different regions quickly. One that marks this convenience is news information or news articles in online content. The ease of accessing news articles is not balanced with the ease of getting the information contained in the news presentation because of the limited time. So we need an automatic summarizing machine that is able to help lay people read the news briefly but directly to the main topic in the article. The automatic text summarizing machine utilizes clustering techniques in grouping sentences that the system considers to have the same meaning based on the weights calculated by the TFIDF algorithm. In an effort to determine an accurate summary, automatic text summation uses the K-Means and K-Medoid algorithms to compare. Testing is done by placing a compression rate of 10%, 20%, 30% on each algorithm. In the results of tests conducted using the BLEU (Billingual Evaluation Understanding) algorithm, K-Means obtained a high precision value at a compression rate of 30% with an average precision value of 45.3%. Whereas K-Medoid gets the highest precision value at 30% compression rate with an average precision value of 44%.