Deep Learning Approaches for Similarity Computation: A Survey
The DOI matches its Crossref record; 121 of the 121 references it cites were found in a registry or an index.
Abstract
The requirement for appropriate ways to measure the similarity between data objects is a common but vital task in various domains, such as data mining, machine learning and so on. Driven by abundant real-world applications, many well-known similarity (distance) metrics are proposed to measure the pairwise similarity of data pairs, e.g., graph edit distance for graphs and dynamic time warping for time series. However, many similarity metrics suffer from the high time complexity. More specifically, most of the well-known similarity metrics often need quadratic time or even much more time to compute the ground truth similarity and some of them are proven to be NP-hard. With the development of deep learning techniques, there is an emerging research trend on the learning for similarity computation on various data types in the field of database (DB) and data mining, which is quite different with the metric learning studies in the machine learning (ML) literature. Specifically, the studies in the ML focus on the learning for semantic similarity in specific tasks, which is implicitly indicated by the training data, on the data in the feature space. While the studies in the DB literature usually consider the learning for well-defined similarity metrics (e.g., graph edit distance) on the data objects (e.g., graphs), such that it can benefit the similarity computation on data in terms of multiple aspects, such as computation time, metric quality and search heuristic, and the learned representation of data can also be naturally fed to downstream tasks. This survey paper provides a comprehensive review of similarity computation learning on several data types, including set, sequence and graph. Moreover, we first classify the learning-based approaches in terms of their learning target into three categories, i.e., similarity learning, cost matrix learning and search heuristic learning. Then we detail some representative approaches for each category on every data type, and analyze some key features that are utilized by these approaches. Finally, we discuss some challenges and future directions towards the learning for similarity learning on these data types.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Conversely, modern deep-learning metrics like LPIPS offer better perceptual alignment but remain computationally prohibitive for real-time, resource-constrained environments.”
From A Survey and Tutorial on Image Quality Assessment with a Contrast-Weighted Structural Similarity Framework (2026) · cites this paper as [32]
In the citing paper
“Deep learning-based similarity [29-34]: Semantic similarity learning uses convolutional neural networks (CNNs) to embed images in a structured feature space. Metrics such as LPIPS (learned perceptual image patch similarity) correlate strongly with human judgment but incur high computational costs, offer low interpretability (a 'black-box' nature), and are vulnerable to adversarial perturbations.”
From A Survey and Tutorial on Image Quality Assessment with a Contrast-Weighted Structural Similarity Framework (2026) · cites this paper as [32]
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
Exploiting Generative AI to Scale up Intelligent Tutoring Systems
Jakubův, J., Chvalovský, K., Goertzel, Z., Kaliszyk, C., Olšák, M., Piotrowski, B., Schulz, S., Suda, M., & Urban, J. (2023) · DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · doi:10.4230/lipics.itp.2023.19 · Cited by 76,658
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Reinforcement Learning: An Introduction
Sutton, R. S., & Barto, A. G. (1998) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.1998.712192 · Cited by 27,225
Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1412.3555 · Cited by 11,008
A Comprehensive Survey on Graph Neural Networks
Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2020.2978386 · Cited by 11,000
node2vec: Scalable Feature Learning for Networks
Grover, A., & Leskovec, J. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939754 · Cited by 10,926
Inductive Representation Learning on Large Graphs
Hamilton, W. L., Ying, R., & Leskovec, J. (2017) · arXiv · doi:10.48550/arxiv.1706.02216 · Cited by 10,639
The graph neural network model
Scarselli, F., Gori, M., Tsoi, A. C., Hagenbuchner, M., & Monfardini, G. (2009) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.2008.2005605 · Cited by 9,582
Research works printed but not linked: 1
- Li, LES3: Learning-based exact set similarity search, 14, 2073
Cited by
Smolen, J. A., Moore, G. E., Perez, N. D., & Wooley, K. L. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2509745122 · Cited by 9
“A secondary model can then use these embeddings for a variety of tasks, including classification, clustering, and object recognition, among others ( 20 ).”
InfluentialAn evaluation review of user similarity metrics in sparse collaborative filtering datasets
Sgardelis, K., Margaris, D., Spiliotopoulos, D., & Vassilakis, C. (2025) · International Journal of Data Science and Analytics · doi:10.1007/s41060-025-00846-4 · Cited by 7
“Deep learning-based approaches for computing user similarity [93, 94] have emerged as a promising development in the RecSys domain, and are currently gaining acceptance.”
Multi-dimensional Semantic Self-learning Hashing for multimedia retrieval
Chen, Y., Long, Y., Yang, Z., & Long, J. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103607 · Cited by 6
SEGMN: A structure-enhanced graph matching network for graph similarity learning
Wang, W., Lu, J., Chen, K., Liu, Z., & Sang, S. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.112530 · Cited by 5
A Hybrid Approach to Typo Correction in Indonesian Documents Using Levenshtein Distance
Santoso, J. T., & Song, Y. (2024) · Journal of Technology Informatics and Engineering · doi:10.51903/jtie.v3i2.184 · Cited by 5
“This dichotomy leaves a gap where a more balanced, hybrid approach could be more effective, particularly in handling the complex morphological structure of Indonesian (Aligon et al., 2014; Laouafi et al., 2022; Yang et al., 2024; Ye et al., 2023).”
BackgroundDeep Probabilistic Graph Matching
Liu, H., Wang, T., Lang, C., Li, Y., & Ling, H. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3578989 · Cited by 5
“However, these methods employ simple and shallow parametric models to control geometric affinities between pairs of matches, and the promotion to the matching accuracy is still limited.”
CharFormer: Character-oriented attention network for string edit distance
Liu, X., Wei, H., Yang, P., & Hu, H. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113482 · Cited by 3
Self-citationGraph Edit Distance Estimation: A New Heuristic and A Holistic Evaluation of Learning-based Methods
Xu, M., & Chang, L. (2025) · Proceedings of the ACM on Management of Data · doi:10.1145/3725304 · Cited by 2
Machine Learning for Graph Data Management and Query Processing
Wang, H., Zhang, Y., & Zhang, W. (2025) · Proceedings of the VLDB Endowment · doi:10.14778/3750601.3750702 · Cited by 2
“Recently, machine learning techniques have been developed for the computation of graph similarity [39] to alleviate the computational costs of this task.”
Self-citationMVPHI: a multi-view learning framework for predicting complex microbial interactions
Xie, Y., Pan, J., Li, D., Wang, Q., Sun, Y., & Wang, S. (2025) · Scientific Reports · doi:10.1038/s41598-025-32359-2 · Cited by 2
Jiang, J., Zuo, Y., & Li, Z. (2026) · Advanced Engineering Informatics · doi:10.1016/j.aei.2026.104675 · Cited by 2
Graph Similarity Learning Based on Learnable Augmentation and Multi-Level Contrastive Learning
Feng, J. G., Guo, Y., & Du, C. (2025) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2025.059610 · Cited by 1
“The second category computes approximate values [11].”
Beyond Locations: A Motion Range-Aware Similarity Join
Li, K., Chen, L., Shang, S., Jensen, C. S., & Kalnis, P. (2025) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3711896.3736855 · Cited by 1
Gao, X. (2025) · Science and Technology of Engineering Chemistry and Environmental Protection · doi:10.61173/bwn0ez36 · Cited by 3
DiveScan: um módulo para Recuperação de dados por Similaridade com Diversidade em Postgres
Lauton, A. J. C., Traina, A. J. M., & Traina, C. (2025) · Simpósio Brasileiro de Bancos de Dados (SBBD) · doi:10.5753/sbbd.2025.247779 · Cited by 1
Schuler, N., Dewald, L., & Graf, J. (2025) · International Conference on Research and Education in Mechatronics (REM) · doi:10.1109/mecatronics-rem67547.2025.11349586 · Cited by 1
“Note that we use the same dataset for both evaluations.”
Yang, J., Wang, X., Ni, Q., Liu, Y., Wang, J., & Wang, F.-Y. (2024) · International Conference on Cyber-Physical Social Intelligence (ICCSI) · doi:10.1109/iccsi62669.2024.10799393 · Cited by 1
“Subsequently, the similarity scores between the query embedding and chunk embeddings within the indexed corpus are computed through dot product similarity, cosine similarity, or other advanced similarity computation methods [56].”
Self-citationHoLoSig: Holistic and Local Representation Learning for Online Signature Verification
de Almeida, J. P. F., Macedo, L. D. A. B., & Freitas, P. G. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-04630-7_25 · Cited by 1
Agaian, S. S., Grigoryan, A. M., & Ayunts, H. (2026) · Information · doi:10.3390/info17070632
Subcellular Region Morphology Reflects Cellular IdentityPreprint
Román, Á., Díaz-Pizarro, A., Pino, N. D. V.-D., Olivera-Gómez, M., Cumplido-Laso, G., Benitez, D. A., Carvajal-González, J. M., & Mulero‐Navarro, S. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.08.14.607889
Architecture of multi-agent systems for generative automatic matching among heterogeneous systems
Batouta, Z. I., Dehbi, R., & Talea, M. (2025) · International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering · doi:10.11591/ijece.v15i2.pp2345-2355
“…methods-such as static identifier-based techniques (SIB) [4]–[6], signature-based techniques (SIG) [7]–[9], similarity-based techniques (SIM) [10]–[15], and custom-specific language techniques (CSL) [16]–[20]-have aimed to match and establish correspondences between different architectural…”
SiamCircle: Trajectory Representation Learning in Free Settings
Nasri, M., Baratchi, M., Koutamanis, A., & Rieffe, C. J. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-91398-3_6
Data Optimization for LLMs: A Survey
Wu, O. (2026) · ACM Transactions on Knowledge Discovery from Data · doi:10.36227/techrxiv.174776562.20873028/v1 · Cited by 4
Yang, 王. W., Shixin, 马. M., Zhan, 李. L., Qingxue, 姚. Y., Chuanyi, 卢. L., Jie, 黄. H., & Xiaogang, 强. Q. (2025) · Acta Optica Sinica · doi:10.3788/aos251369
Interactive hypergraph visual analytics for exploring large and complex image collections
Gisolf, F., Geradts, Z., & Worring, M. (2026) · Information Visualization · doi:10.1177/14738716261459480
GEDAN: Learning the Edit Costs for Graph Edit DistancePreprint
Leonardi, F., Orsi, M., Reymond, J.-L., & Riesen, K. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.03111
“This contradiction was recently pointed out by Yang et al. [19], who proposed strategies such as transfer learning and few-shot learning to mitigate the problem.”
Shendi, T. A., Darwish, S. M., & Younes, A. (2025) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-3-032-07336-5_21
Similarity Reasoning in Design Engineering Ontologies: A Conceptual Framework
BURDULEA, I.-L., Starchenko, S., Lünnemann, P., & Lindow, K. (2025) · International Conference on Computational Intelligence and Knowledge Economy (ICCIKE) · doi:10.1109/iccike67021.2025.11318291
APER: An Efficient and Privacy-Preserving Scheme for E-Health Recommendations
Wang, J., Yuan, W., Deng, X., Zhang, Q., Fan, J., & Xue, Y. (2025) · IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom) · doi:10.1109/trustcom66490.2025.00089
Lu, Y., Li, Y., Zhong, Q., Zhong, J., Yu, Y., & Zhang, Y. (2026) · Aerospace · doi:10.3390/aerospace13040314
Self-citationBehavioral Analysis of Semantic Similarity Metrics under Transformer-Based Representations
Pradana, M. G., Irzavika, N., Maulana, N., Karenina, S. A., & Rabbani, S. A. (2026) · Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) · doi:10.29207/resti.v10i2.7490
Exploring Approaches for Curve Similarity: A Comprehensive Review
Mishra, S., & Tiwari, N. (2026) · International Journal of Mathematical Sciences and Computing · doi:10.5815/ijmsc.2026.03.07
Robust Spatial-Temporal Similar Trajectory Search via Structure-Enhanced Domain-Invariant Learning
Han, X., Zhou, Y., Ma, C., Song, L., Gan, X., & Shang, X. (2026) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde65706.2026.00239
“Similar trajectory search aims to retrieve trajectories similar to a given query, which is beneficial for a wide range of applications, such as route recommendation, traffic pattern analysis and location-based services, e.g., ride-sharing and logistics optimization [1]–[5].”
Pixel-level maize aflatoxin B1 spectral detection using self-supervised learning
Guo, W., Zhu, H., & Han, Z. (2026) · Microchemical Journal · doi:10.1016/j.microc.2026.119614
PRIDE: Diversifying Similarity Queries in RDBMSs Using Pivots
Lauton, A. J. C., Oliveira, W. D., Traina, A. J. M., & Traina, C. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-39820-8_22
A Multi-Scale State Space Representation Method for Spacecraft Component Matching
Liu, P., Yang, Z., Liu, C., & Li, C. (2026) · International Conference on Computer Network, Electronic and Automation (ICCNEA) · doi:10.1109/iccnea70678.2026.11722269
“Classical retrieval methods include alignment-based comparison, distance metrics, and learned time-series embeddings [8]–[12].”
Metadata
- Authors
- Yang, P., Wang, H., Yang, J., Qian, Z., Zhang, Y., & Lin, X.
- Affiliations
- Zhejiang Lab (CN)
- Zhejiang Gongshang University (CN)
- Guangzhou University (CN)
- Alibaba Group (China) (CN)
- Shanghai Jiao Tong University (CN)
- Venue
- IEEE Transactions on Knowledge and Data Engineering
- Published
- 2024-12
- Volume
- 36
- Issue
- 12
- Pages
- 7893-7912
- ISSN
- 1041-4347 (ISSN-L)
- 1558-2191
- 2326-3865
- OpenAlex
- W4400275701
- Cited by
- 28
- References
- 117
- Citations by year
- 2026: 7 · 2025: 18 · 2024: 3
- FWCI
- 8.52
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Knowledge and Data Engineering (publishedVersion)
- Funders
- Major Key Project of PCL (PCL2022A03)
- Basic and Applied Basic Research Foundation of Guangdong Province (2023A1515011655)
- Guangzhou Research Foundation (2023A03J0115)
- Topics
- Physical Sciences › Computer Science › Information Systems › Data Quality and Management
- Physical Sciences › Computer Science › Information Systems › Advanced Database Systems and Queries
- Physical Sciences › Computer Science › Information Systems › Data Management and Algorithms
- Keywords
- similarity measurement, metric learning, graph edit distance, dynamic time warping, set similarity, sequence similarity, graph similarity, pairwise similarity, NP-hardness, computational complexity
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.