Investigation on Handwritten Mathematical Symbol Recognition Based on the Combination of CNN and KNN Method

DOI

10.2991/978-94-6463-300-9_66How to use a DOI?

Keywords

CNN; KNN; Machine learning; Handwritten mathematical symbols recognition

Abstract

Recognizing handwritten mathematical symbols presents a significant obstacle due to the inherent variability in individuals’ writing styles. In order to enhance the accuracy of symbol recognition, this scholarly article introduces a pioneering methodology that synergistically merges the capabilities of the Convolutional Neural Network (CNN) and the K-nearest Neighbors algorithm (KNN). This approach endeavors to leverage the respective advantages offered by both CNN and KNN, with the ultimate objective of advancing the accuracy of symbol identification. Primarily, the CNN model undergoes multiple rounds of training to augment its feature extraction capabilities. Subsequently, the extracted features are employed for training and classification predictions within the KNN framework, yielding the final predicted results. To evaluate the performance of this approach, tests are conducted on the Handwritten Math Symbols dataset from Kaggle, and comparisons are made with methods that solely employ CNN or KNN. All three models are evaluated using identical training and testing datasets. The results demonstrate that the combined CNN and KNN approach outperforms in various performance indicators, achieving an ultimate accuracy of 98.7%. This evidences the superior performance of this method in the task of handwritten mathematical symbol recognition.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yisong Zhang
PY  - 2023
DA  - 2023/11/27
TI  - Investigation on Handwritten Mathematical Symbol Recognition Based on the Combination of CNN and KNN Method
BT  - Proceedings of the 2023 International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2023)
PB  - Atlantis Press
SP  - 638
EP  - 646
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-300-9_66
DO  - 10.2991/978-94-6463-300-9_66
ID  - Zhang2023
ER  -