Skripsi
CLUSTERING ARTIKEL ILMIAH BERBAHASA INDONESIA MENGGUNAKAN ALGORITMA PARTITIONING AROUND MEDOIDS
A scientific article is a type of written work by a researcher after successfully conducting research in a particular field, which is arranged systematically based on writing rules. Each scientific article contains important information from various scientific fields, one of which is computer science. In this research, an automatic clustering software for scientific articles in Indonesian was built, where the data used was the abstract of each computer science scientific article. Clustering is a method for grouping objects into several clusters so that each cluster contains objects that are similar based on certain categories. In this research, the clustering process was carried out to find out words that often and rarely appear in computer science scientific articles. The methods used in software development begin with the text preprocessing process, word weighting with TF-IDF, searching for the best k value with Silhouette Score, and the clustering process with the Partitioning Around Medoids algorithm, better known as k-Medoids. The experimental results of the software that has been built were the formation of 10 optimal clusters in the first experiment using the best k value (k=10 with a Silhouette Score of 0.7123), and the formation of 1 cluster in the second experiment using k=2 (with a Silhouette Score of 0). This software also has good performance, as seen in every experimental process carried out
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2407002578 | T143471 | T1434712024 | Central Library (REFERENCES) | Available but not for loan - Not for Loan |
No other version available