Page 78 - 《应用声学》2021年第5期

722 2021 年 9 月

于最小可检测阈6 dB 的数据上，未来将进一步研究 IEEE 36th International Conference on Data Engineer-
该算法在更低可检测信噪比数据集上的不平衡分 ing. IEEE Computer Society, 2020: 841–851.
[7] Han J, Kamber M, Pei J. 数据挖掘: 概念与技术 [M]. 范明,
类效果。 孟小峰, 译. 第三版. 北京: 机械工业出版社, 2019: 250–251.
[8] Núez H, Gonzalez-Abril L, Angulo C. Improving SVM
classification on imbalanced datasets by introducing a new
参 考 文 献 bias[J]. Journal of Classification, 2017, 34(3): 427–443.
[9] Vanhoeyveld J, Martens D. Imbalanced classification in
sparse and large behaviour datasets[J]. Data Mining &
[1] Berg H, Hjelmervik K T. Classification of anti-submarine Knowledge Discovery, 2018, 32(1): 1–58.
warfare sonar targets using a deep neural network[C]. [10] Lin Y, Yoonkyung L, Grace W. Support vector machines
OCEANS Marine Technology Society. IEEE Charleston, for classification in nonstandard situations[J]. Machine
2018: 1–5. Learning, 2002, 46(1/2/3): 191–202.
[2] Berg H, Hjelmervik K T. A comparison of different ma- [11] Zheng E, Li P, Song Z. Cost sensitive support vector ma-
chine learning algorithms for automatic classification of chines[J]. Control & Decision, 2006, 21(4): 473–476.
sonar targets[C]. OCEANS Marine Technology Society. [12] Liu N, Qi E, Xu M, et al. A novel intelligent classifi-
IEEE Monterey, 2016: 1–8. cation model for breast cancer diagnosis[J]. Information
[3] Stender D H, Hjelmervik K T, Berg H, et al. Sensitivity Processing & Management, 2019, 56(3): 609–623.
to target behavior in automatic classification on kinematic [13] 于化龙. 类别不平衡学习: 理论与算法 [M]. 北京: 清华大学
track features[C]. OCEANS Marine Technology Society. 出版社, 2017: 4–5.
IEEE Kobe, 2018: 1–5. [14] Zhang T. Statistical behavior and consistency of classifi-
[4] Stender D H, Hjelmervik K T, Berg H, et al. The classifi- cation methods based on convex risk minimization[J]. The
cation performance of signal-to-noise ratio and kinematic Annals of Statistics, 2004, 32(1): 56–85.
features in varying environments[C]. OCEANS. IEEE Ab- [15] Lin H T, Lin C J, Weng R C. A note on Platt’s proba-
erdeen, 2017: 1–5. bilistic outputs for support vector machines[J]. Machine
[5] 赵楠, 张小芳, 张利军. 不平衡数据分类研究综述 [J]. 计算机 Learning, 2007, 68(3): 267–276.
科学, 2018, 45(6A): 22–27. [16] Székely G J, Rizzo M L. Energy statistics: a class of statis-
[6] Liu Z, Cao W, Gao Z, et al. Self-paced ensemble tics based on distances[J]. Journal of Statistical Planning
for Highly imbalanced massive data classification[C]. and Inference, 2013, 143(8): 1249–1272.