A Learning Vector Quantization Approach to Handwritten Mandarin Numeral Recognition
Abstract
Numbers 1 to 10 in Mandarin are also studied in the Mandarin language learning process as basic numbers. Mandarin numbers have a different shape from Arabic numbers and Roman numerals. So it is necessary to recognize the pattern of mandarin numbers to help the learning process of mandarin. Therefore, the purpose of this research is to build an application that applies the Learning Vector Quantization method for handwriting pattern recognition of Mandarin numbers. System testing methods used are Black Box and Confusion Matrix for accuracy testing methods. The application that has been made produces an accuracy of 92.80% with a total of 250 test data. Keyword: Learning Vector Quantization, Pattern Recognition, Mandarin Numbers.
Fields of Study & Subjects
Handwritten Text Recognition Techniques
Data Mining and Machine Learning Applications
Vehicle License Plate Recognition
Publication details
- Journal / source
- Jurnal Ilmiah SINUS
- Volume
- 24
- Issue
- 1
- Pages
- 35–44
- Publication date
- 6 February 2026
- Language
- EN