K-nearest neighbor performance for Nusantara scripts image transliteration
Open Access
- 13 March 2020
- journal article
- Published by Institute of Research and Community Services Diponegoro University (LPPM UNDIP) in Jurnal Teknologi dan Sistem Komputer
- Vol. 8 (2), 150-156
- https://doi.org/10.14710/jtsiskom.8.2.2020.150-156
Abstract
The concept of classification using the k-nearest neighbor (KNN) method is simple, easy to understand, and easy to be implemented in the system. The main challenge in classification with KNN is determining the proximity measure of an object and how to make a compact reference class. This paper studied the implementation of the KNN for the automatic transliteration of Javanese, Sundanese, and Bataknese script images into Roman script. The study used the KNN algorithm with the number k set to 1, 3, 5, 7, and 9. Tests used the image dataset of 2520 data. With the 3-fold and 10-fold cross-validation, the results exposed the accuracy differences if the area of the extracted image, the number of neighbors in the classification, and the number of data training were different.Keywords
Funding Information
- Universitas Sanata Dharma (042/LPPM USD/V/2019)
This publication has 9 references indexed in Scilit:
- Comparison of distance measurement on k-nearest neighbour in textual data classificationJurnal Teknologi dan Sistem Komputer, 2019
- Human action recognition using a corners and blob detector with different classification methodsIOP Conference Series: Materials Science and Engineering, 2019
- Analysis and Impact Evaluation of Missing Data Imputation in Day-ahead PV Generation ForecastingApplied Sciences, 2019
- Automatic Recognition of the NIK in Electronic KTPPublished by European Alliance for Innovation n.o. ,2019
- Yoruba Handwritten Character Recognition using Freeman Chain Code and K-Nearest Neighbor ClassifierJurnal Teknologi dan Sistem Komputer, 2018
- Using K-Nearest Neighbor in Optical Character RecognitionComTech: Computer, Mathematics and Engineering Applications, 2016
- Voting over Multiple Condensed Nearest NeighborsPublished by Springer Science and Business Media LLC ,1997
- A bootstrap technique for nearest neighbor classifier designIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997
- Nearest neighbor pattern classificationIEEE Transactions on Information Theory, 1967