As an important material carrier of urban space, road network is considered as the "fingerprint" of a city. Its model reflects the modernization process of a city. Extracting and analyzing the structural model of urban road network is helpful to understand urban development and build a better urban form. In order to solve the shortcomings of the existing research on pattern recognition methods of urban road network, this paper discusses the pattern classification of urban road network, the pattern recognition of road network distribution and the support vector machine, discusses the selection and establishment of road network samples and the recognition parameters of road network patterns, and discusses the design of the pattern recognition model of urban road network. Finally, the recognition model designed in this paper is compared with CNN, RF and GBDT. The experimental results show that the recall rate, accuracy rate and recall rate of SVM urban road network pattern recognition reach 97.8%, which is superior to the other three recognition methods. Therefore, it is verified that the pattern recognition method of urban road network based on support vector machine algorithm has high practical value.

Feng Liu. Pattern Recognition Method of Urban Road Network Based on Support Vector Machine Algorithm. Machine Learning Theory and Practice (2022), Vol. 3, Issue 3: 69-77. https://doi.org/10.38007/ML.2022.030309.

[1] Bohua, HUANG, Wei, et al. Hierarchical Area Partitioning Method of Urban Road Networks Matching. Journal of Geodesy and Geoinformation Science, 2019, v.2(03):57-69.

[2] Williams, Julie, C, et al. Soluble CD14, CD163, and CD27 biomarkers distinguish ART‐suppressed youth living with HIV from healthy controls. Journal of Leukocyte Biology: An Official Publication of the Reticuloendothelial Society, 2018, 103(4):671-680. https://doi.org/10.1002/JLB.3A0717-294RR

[3] Roy J, Han J, Weeks A, et al. P.138 A small RNA signature from extracellular vesicles in patient plasma correlates with recurrence or progression of high-grade gliomas. Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques, 2022, 49(s1):S43-S43. https://doi.org/10.1017/cjn.2022.222

[4] Nino, Kopaleishvili. Perception of Myth in Print and Online Media. Language and Semiotic Studies, 2020, v.6(01):35-45.

[5] Chiu A P. When Rifles Need a Makeover: SARCO Inc.'s Reproduction G43 Magazine. Small arms review, 2019, 23(8):44-47.

[6] Ewelina, Wawryk-Gawda, Beata, et al. Dietary supplements – consumer assessment based on questionnaire survey. Przeglad epidemiologiczny, 2018, 72(1):111-120.

[7] Tsertsvadze G. The Probabilistic Model of Canonically Conjugate Fuzzy Subsets. Bulletin of the Georgian Academy of Sciences, 2018, 12(4):53-58.

[8] Gitonga, Njeru. Roads of Success Ethiopia's new extensive road network to boost industrialization development. ChinAfrica, 2019, v.11(03):34-35.

[9] Memon I. Distance and clustering‐based energy‐efficient pseudonyms changing strategy over road network. International Journal of Communication Systems, 2018, 31(11):e3704.1-e3704.22. https://doi.org/10.1002/dac.3704

[10] Dhanpat J, Higginson A, Brooks K. Estimation of the Effect of Bio-Admixtures on Concrete Workability Using Linear Regression and Support Vector Machines. IFAC-PapersOnLine, 2021, 54( 21):133-138. https://doi.org/10.1016/j.ifacol.2021.12.023

[11] Kahl M, Kroll A. Extending Regularized Least Squares Support Vector Machines for Order Selection of Dynamical Takagi-Sugeno Models. IFAC-PapersOnLine, 2020, 53( 2):1182-1187. https://doi.org/10.1016/j.ifacol.2020.12.1331

[12] Kantavat P, Kijsirikul B, Songsiri P, et al. Efficient Decision Trees for Multi-Class Support Vector Machines Using Entropy and Generalization Error Estimation. International Journal of Applied Mathematics & Computer Science, 2018, 28(4):705-717. https://doi.org/10.2478/amcs-2018-0054

[13] Mathew J, Pang C K, Luo M, et al. Classification of Imbalanced Data by Oversampling in Kernel Space of Support Vector Machines. Neural Networks and Learning Systems, IEEE Transactions on, 2018, 29(9):4065-4076. https://doi.org/10.1109/TNNLS.2017.2751612

[14] Kim, Sungdo, Mun, et al. Data depth based support vector machines for predicting corporate bankruptcy. Applied Intelligence: The International Journal of Artificial Intelligence, Neural Networks, and Complex Problem-Solving Technologies, 2018, 48(3):791-804. https://doi.org/10.1007/s10489-017-1011-3

[15] Mohsen, Farhang, Ali, et al. Automatic modulation recognition for DVB-S2 using pairwise support vector machines. International journal of autonomous and adaptive communications systems: IJAACS, 2018, 11(2):144-153. https://doi.org/10.1504/IJAACS.2018.092020

[16] Silvina, Niell, Florencia, et al. Beehives biomonitor pesticides in agroecosystems: Simple chemical and biological indicators evaluation using Support Vector Machines (SVM). Ecological indicators: Integrating, monitoring, assessment and management, 2018, 91(Aug.):149-154. https://doi.org/10.1016/j.ecolind.2018.03.028

[17] Hossain M Z, Gedeon T, Sankaranarayana R. Using Temporal Features of Observers' Physiological Measures to Distinguish Between Genuine and Fake Smiles. IEEE Transactions on Affective Computing, 2020, 11(1):163-173. https://doi.org/10.1109/TAFFC.2018.2878029

[18] Williams, Julie, C, et al. Soluble CD14, CD163, and CD27 biomarkers distinguish ART‐suppressed youth living with HIV from healthy controls. Journal of Leukocyte Biology: An Official Publication of the Reticuloendothelial Society, 2018, 103(4):671-680. https://doi.org/10.1002/JLB.3A0717-294RR