Education, Science, Technology, Innovation and Life
Open Access
Sign In

Siamese Network-Based Text Similarity Algorithm Research

Download as PDF

DOI: 10.23977/jaip.2024.070315 | Downloads: 18 | Views: 1078

Author(s)

Junhong Chen 1,2, Kaihui Peng 3

Affiliation(s)

1 School of Software Engineering, South China University of Technology, Guangzhou, China
2 LeiHuo Studio, NetEase, Hangzhou, China
3 Faculty of Business and Economics, University of Malaya, Kuala Lumpur, Malaysia

Corresponding Author

Junhong Chen

ABSTRACT

This paper proposes a text similarity calculation model based on multi-scale convolutional neural networks and attention mechanisms. The model is capable of extracting information at different granularities within the text, enabling it to learn from multiple layers of information and thereby improving the accuracy of the text similarity calculation task. After training, the model can generate sentence vectors suitable for cosine similarity computation, which allows the model to pre-generate vectors for the text in the repository. During actual retrieval, only the sentence vector of the text to be searched is needed to calculate similarity with the pre-generated vectors in the repository.

KEYWORDS

Text similarity calculation; Attention mechanism; Pre-trained model

CITE THIS PAPER

Junhong Chen, Kaihui Peng, Siamese Network-Based Text Similarity Algorithm Research. Journal of Artificial Intelligence Practice (2024) Vol. 7: 123-131. DOI: http://dx.doi.org/10.23977/jaip.2024.070315.

REFERENCES

[1] Mikolov T, Chen K, Corrado G, et al. Efficient estimation of word representations in vector space [J]. arXiv preprint arXiv:1301.3781, 2013.
[2] Ramos J. Using tf-idf to determine word relevance in document queries[C]//Proceedings of the first instructional conference on machine learning. 2003, 242(1): 29-48. 
[3] Cui Y, Che W, Liu T, et al. Pre-training with whole word masking for chinese bert[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021, 29: 3504-3514.
[4] Koch G, Zemel R, Salakhutdinov R. Siamese neural networks for one-shot image recognition[C]//ICML deep learning workshop. 2015, 2(1): 1-30.
[5] Cai Z, Fan Q, Feris R S, et al. A unified multi-scale deep convolutional neural network for fast object detection[C]//Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part IV 14. Springer International Publishing, 2016: 354-370.
[6] Vaswani A. Attention is all you need [J]. Advances in Neural Information Processing Systems, 2017.
[7] Huang A. Similarity measures for text document clustering[C]//Proceedings of the sixth new zealand computer science research student conference (NZCSRSC2008), Christchurch, New Zealand. 2008, 4: 9-56.
[8] Mairal J, Koniusz P, Harchaoui Z, et al. Convolutional kernel networks[J]. Advances in neural information processing systems, 2014, 27.
[9] Niu Z, Zhong G, Yu H. A review on the attention mechanism of deep learning[J]. Neurocomputing, 2021, 452: 48-62.
[10] Yu D, Wang H, Chen P, et al. Mixed pooling for convolutional neural networks[C]//Rough Sets and Knowledge Technology: 9th International Conference, RSKT 2014, Shanghai, China, October 24-26, 2014, Proceedings 9. Springer International Publishing, 2014: 364-375.
[11] Liu X, Chen Q, Deng C, et al. Lcqmc: A large-scale chinese question matching corpus[C]//Proceedings of the 27th International Conference on Computational Linguistics. 2018: 1952-1962.
[12] Chen J, Chen Q, Liu X, et al. The bq corpus: A large-scale domain-specific chinese corpus for sentence semantic equivalence identification[C]//Proceedings of the 2018 conference on empirical methods in natural language processing. 2018: 4946-4951.
[13] Joulin A, Grave E, Bojanowski P, et al. Bag of tricks for efficient text classification[J]. arXiv preprint arXiv:1607.01759, 2016.
[14] Reimers N. Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks[J]. arXiv preprint arXiv:1908.10084, 2019.

Downloads: 15127
Visits: 485220

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.