Welcome to Francis Academic Press

Academic Journal of Computing & Information Science, 2023, 6(11); doi: 10.25236/AJCIS.2023.061105.

Difficulty Level Classification of Sudoku Puzzles Based on Convolutional Neural Network


Xuan Wei

Corresponding Author:
Xuan Wei

Department of Computer Science, The University of Manchester, Manchester, The United Kingdom


Sudoku is a classic logic puzzle that many people love to play. Dividing the difficulty of Sudoku puzzles helps provide Sudokus with different levels suitable for new or skilled Sudoku players. This paper proposes a model using convolutional neural networks to distinguish difficulty levels of Sudoku puzzles. Firstly, this paper uses traditional depth-first search algorithms to measure the solving steps of Sudoku, thereby labelling the difficulty of Sudoku puzzles. Then, these difficulty-labelled Sudoku data are used as training data to enable the convolutional neural network-based model to distinguish Sudokus with difficulty levels. Finally, this neural network model can correctly classify approximately 80% of the difficulty levels in testing Sudoku datasets.


Deep learning, Convolutional neural network, Supervised learning, Sudoku puzzle, Depth-first search algorithm

Cite This Paper

Xuan Wei. Difficulty Level Classification of Sudoku Puzzles Based on Convolutional Neural Network. Academic Journal of Computing & Information Science (2023), Vol. 6, Issue 11: 35-39. https://doi.org/10.25236/AJCIS.2023.061105.


[1] Job D, Paul V. Recursive backtracking for solving 9* 9 Sudoku puzzle[J]. Bonfring International Journal of Data Mining, 2016, 6(1): 7-9. 

[2] LeCun Y, Bengio Y, Hinton G. Deep learning[J]. nature, 2015, 521(7553): 436-444.

[3] Vamsi K S, Gangadharabhotla S, Sai V S H. A Deep Learning approach to solve sudoku puzzle [C]//2021 5th International Conference on Intelligent Computing and Control Systems (ICICCS). IEEE, 2021: 1175-1179. 

[4] Gu J, Wang Z, Kuen J, et al. Recent advances in convolutional neural networks[J]. Pattern recognition, 2018, 77: 354-377. 

[5] Rawat W, Wang Z. Deep convolutional neural networks for image classification: A comprehensive review [J]. Neural computation, 2017, 29(9): 2352-2449.

[6] Noriega L. Multilayer perceptron tutorial[J]. School of Computing. Staffordshire University, 2005, 4(5): 444.

[7] Gholamalinezhad H, Khosravi H. Pooling methods in deep neural networks, a review[J]. arXiv preprint arXiv:2009.07485, 2020. 

[8] Jordan M I, Mitchell T M. Machine learning: Trends, perspectives, and prospects[J]. Science, 2015, 349(6245): 255-260.

[9] Kingma D P, Ba J. Adam: A method for stochastic optimization[J]. arXiv preprint arXiv:1412.6980, 2014.

[10] Ruder S. An overview of gradient descent optimization algorithms[J]. arXiv preprint arXiv:1609.04747, 2016.

[11] Pan S J, Yang Q. A survey on transfer learning[J]. IEEE Transactions on knowledge and data engineering, 2009, 22(10): 1345-1359.