PENGATEGORIAN ARTIKEL BERITA KAMPUS BERBAHASA INDONESIA PADA PORTAL BERITA KAMPUS DENGAN MENGGUNAKAN ALGORITMA BISECTING K-MEANS
| Main Authors: | , Putu Bagus Susastra, , Ir. P. Insap Santosa, M.Sc., Ph.D |
|---|---|
| Format: | Thesis NonPeerReviewed |
| Terbitan: |
[Yogyakarta] : Universitas Gadjah Mada
, 2013
|
| Subjects: | |
| Online Access: |
https://repository.ugm.ac.id/122774/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=62880 |
Daftar Isi:
- Campbuzz is campus news portal that obtained the news from universities RSS feed so that the amount of managed articles is very big and there is no category on news articles. It will certainly increase the workload of Campbuzz admin so is needed system that can manage the articles in the Campbuzz database to ease Campbuzz admin workload. Clustering is unsupervised learning techniques are used to determine the groups (clusters) of a set of large number data. One of algorithm that used for clustering is bisecting K-means, the improvement of K-means algorithm. Text clustering based on content of news articles can be a solution to the problem faced by Campbuzz admin in case arrangements article in Campbuzz database. Development systems that can perform text clustering begins with a literature study relating to the text preprocessing, document representation techniques, clustering techniques, clustering algorithm and tools used for the clustering process. The result of this research is 20 groups news of 210 campus news article samples. The average value of IST from 20 clusters is 0.590013058. Each cluster is represented by 3 words.