We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Perangkingan Dokumen Berbahasa Arab Berdasarkan Query dengan Metode Klasifikasi Naïve Bayes dan KNearest Neighbor.
- Authors
Kiftiyani, Usfita; Suprapto; Yudistira, Novanto
- Abstract
Research about document ranking on information retrieval is now easy to find, this is related to scientific developments in the field of extracting information that is moving very fast. However, research that used Arabic documents as an object it is still limited. Due to the limited use of Arabic documents for research in the field of extracting information, the author tries to take a simple approach, by implementing the Naïve Bayes and the k-Nearest Neighbor (k-NN) classification method. The purpose of this study was to determine whether the classification methods, especially Naïve Bayes and k-NN, can be used to rank, and also compare the accuracy of the two methods. Based on this research, it was found that the ranking with the classification method can be done with the accuracy level of the Naïve Bayes method is better than the k-NN method with an average F1 Measure value reaching 72%, the average value of precision is 75%, and the average recall value reaches 80%. Meanwhile, the results of the k-NN method showed that the average value of F1 Measure reached 70%, the average value of precision was 76%, and the average recall value reached 79%. However, this research is still lacking in terms of the technique used, which is by eliminating the stemming process. So the authors provide suggestions for further research so that the stemming process can be carried out and using a newer ranking method.
- Subjects
INFORMATION retrieval; CLASSIFICATION; SCIENTIFIC development; ARITHMETIC mean; NEIGHBORS
- Publication
Techno.com, 2020, Vol 19, Issue 4, p321
- ISSN
1412-2693
- Publication type
Article
- DOI
10.33633/tc.v19i4.3939