Abstract
The choice of the appropriate method in the classification task is most often a problem related to the adaptation of the input data to the classifier. However, adaptation alone does not result in high classification scores. In this paper, we present a comparison of two artificial intelligence methods for recognizing and classifying the handwriting of digits. The study was based on the popular MNIST database, and we dug up algorithms such as K-nearest neighbors and also neural networks to conduct the study. The paper presents mathematical models of selected tools and selected network architecture. Then, the results of the research carried out in order to choose a more accurate character classification technique are presented. For the purpose of verification, the accuracy metric and the analysis using the error matrix were used. Article also includes analysis of different variables to used methods, like metrics (Euclidean, Manhattan and Chebyshev) or hyperparameter k.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Cui, P., Athey, S.: Stable learning establishes some common ground between causal inference and machine learning. Nat. Mach. Intell. 4, 110–115 (2022)
Ke, Q., Siłka, J., Wieczorek, M., Bai, Z., Woźniak, M.: Deep neural network heuristic hiearchization for cooperative intelligent transportation fleet management. IEEE Trans. Intell. Transp. Syst. 23, 16752–16762 (2022)
Esterhuizen, J.A., Goldsmith, B.R., Linic, S.: Interpretable machine learning for knowledge generation in heterogeneous catalysis. Nat. Catalysis 5, 175–184 (2022)
Połap, D., Woźniak, M.: A hybridization of distributed policy and heuristic augmentation for improving federated learning approach. Neural Netw. 146, 130–140 (2022)
Rajeshkumar, G., et al.: Smart office automation via faster RCNN based face recognition and internet of things. Measur.: Sens. 27, 100719 (2023)
Siłka, J., Wieczorek, M., Woźniak, M.: Recurrent neural network model for high-speed train vibration prediction from time series. Neural Comput. Appl. 34, 13305–13318 (2022)
de Andrés-Sánchez, J.: A systematic review of the interactions of fuzzy set theory and option pricing. Expert Syst. Appl., 119868 (2023)
Woźniak, M., Zielonka, A., Sikora, A., Piran, M.J., Alamri, A.: 6g-enabled IoT home environment control using fuzzy rules. IEEE Internet Things J. 8, 5442–5452 (2020)
Abualigah, L., et al.: Meta-heuristic optimization algorithms for solving real-world mechanical engineering design problems: a comprehensive survey, applications, comparative analysis, and results. Neural Comput. Appl., 1–30 (2022)
Sikora,A., Zielonka, A., Woźniak, M.: Heuristic optimization of 18-pulse rectifier system. In: 2021 IEEE Congress on Evolutionary Computation (CEC), pp. 673–680. IEEE (2021)
Marewski, J.N., Gigerenzer, G.: Heuristic decision making in medicine. Dial. Clin. Neurosci. (2022)
Połap, D., Srivastava, G.: Neural image reconstruction using a heuristic validation mechanism. Neural Comput. Appl. 33, 10787–10797 (2021)
Prokop, K., Połap, D., Srivastava, G., Lin, J.C.-W.: Blockchain-based federated learning with checksums to increase security in internet of things solutions. J. Ambient Intell. Human. Comput., 1–10 (2022)
Kumar, S., Sahoo, S., Lim, W.M., Kraus, S., Bamel, U.: Fuzzy-set qualitative comparative analysis (FSQCA) in business and management research: a contemporary overview. Technol. Forecast. Soc. Chang. 178, 121599 (2022)
Haseeb, J., Mansoori, M., Hirose, Y., Al-Sahaf, H., Welch, I.: Autoencoder-based feature construction for iot attacks clustering. Futur. Gener. Comput. Syst. 127, 487–502 (2022)
Zhang, J., Li, Y., Xiong, H., Dou, D., Miao, C., Zhang, D.: Handgest: Hierarchical sensing for robust-in-the-air handwriting recognition with commodity wifi devices. IEEE Internet Things J. 9, 19529–19544 (2022)
Acknowledgments
We would like to express our gratitude to the creators of MNIST for making this valuable data available on Kaggle. The efforts of the National Institute of Standards and Technology in collecting and curating this dataset are greatly appreciated.
• MNIST Dataset on Kaggle: https://www.kaggle.com/datasets/crawford/emnist.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Koman, W., Małecki, K. (2024). Comparison of kNN Classifier and Simple Neural Network in Handwritten Digit Recognition Using MNIST Database. In: Lopata, A., Gudonienė, D., Butkienė, R. (eds) Information and Software Technologies. ICIST 2023. Communications in Computer and Information Science, vol 1979. Springer, Cham. https://doi.org/10.1007/978-3-031-48981-5_21
Download citation
DOI: https://doi.org/10.1007/978-3-031-48981-5_21
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-48980-8
Online ISBN: 978-3-031-48981-5
eBook Packages: Computer ScienceComputer Science (R0)