Academic Journal of Mathematical Sciences, 2023, 4(3); doi: 10.25236/AJMS.2023.040310.
Mingzhe Li
Deerfield Academy, Franklin County, Massachusetts, United States of America
Natural language processing (NLP) has greatly advanced in machine learning, but math education software lacks AI integration for solving math word problems in English. We propose using the BertGen pre-trained Transformer model, along with the MAWPS dataset augmented by our dataset augmenter. The Transformer model, with its multi-head attention mechanisms, excels at capturing long-range dependencies and referential relationships, crucial for math word problems at the primary school level. Our accuracy tests and performance on different datasets validate the effectiveness and generalizability of our approach. Moreover, our augmented dataset outperforms smaller unaugmented datasets, while maintaining diversity. The math word problem augmenter can be adapted for other math problem sets, supporting future research in the field.
Math word problem, Natural language processing, Transformer model
Mingzhe Li. Analysis of Elementary Math Word Problems Based on AI Deep Learning. Academic Journal of Mathematical Sciences (2023) Vol. 4, Issue 3: 60-69. https://doi.org/10.25236/AJMS.2023.040310.
[1] Kushman, N., Artzi, Y., Zettlemoyer, L., Barzilay, R. (2014, June). Learning to automatically solve algebra word problems. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 271-281).
[2] Chatterjee, O., Waikar, A., Kumar, V., Ramakrishnan, G., Arya, K. (2021). A weakly supervised model for solving math word problems. arXiv preprint arXiv:2104.06722.
[3] Cheng Y , Li B .Computer data analysis and processing based on character recognition and deep learning technology[C]//ICASIT 2020: 2020 International Conference on Aviation Safety and Information Technology.2020.DOI:10.1145/3434581.3434727.
[4] Shi, S., Wang, Y., Lin, C. Y., Liu, X., Rui, Y. (2015, September). Automatically solving number word problems by semantic parsing and reasoning. In Proceedings of the 2015 conference on empirical methods in natural language processing (pp. 1132-1142).
[5] Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30.
[6] Zhao, W., Shang, M., Liu, Y., Wang, L., Liu, J. (2020). Ape210k: A large-scale and template-rich dataset of math word problems. arXiv preprint arXiv: 2009.11506.
[7] Wang, Y., Liu, X., Shi, S. (2017, September). Deep neural solver for math word problems. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing (pp. 845-854).
[8] Koncel-Kedziorski, R., Roy, S., Amini, A., Kushman, N., Hajishirzi, H. (2016, June). MAWPS: A math word problem repository. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (pp. 1152-1157).
[9] Wang, L., Zhang, D., Gao, L., Song, J., Guo, L., Shen, H. T. (2018). MathDQN: Solving Arithmetic Word Problems via Deep Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 32(1). https://doi.org/10.1609/aaai.v32i1.11981
[10] Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.