基于SVM的食双星光变曲线自动分类算法
|
摘要
提出一种基于机器学习的食双星光变曲线自动分类算法。首先对数据进行预处理,将食双星光变曲线数据归一化,并通过滤波/插值降低噪声;其次使用快速傅里叶变换提取频率信号作为特征向量;最后利用特征向量训练支持向量机获得自动分类模型。使用Python实现算法并抓取CALEB和GCVS数据验证,分析特征向量、支持向量机核函数与惩罚系数对分类正确率的影响,优化后所得分类模型正确率达到92.8%(训练集)和89.0%(测试集),最后使用所得分类模型对第3方数据进行分类,正确率为88.8%,结果证明提出的分类算法的有效性。
This paper proposes an automatic classification algorithm for light curves of eclipsing binary stars based on machine learning. At first the algorithm normalizes the light curves and performs filtering/interpolation to reduce the noise effect in the preprocessing stage, then the Fourier coefficients, which are extracted by FFT from the light curves, are used as the feature vector to train SVM and a classification model is obtained. We implement this algorithm with Python and use the data captured from CALEB/GCVS to validate and discuss the effect of the featured vector, the SVM kernel function and the penalty coefficient on the classifying accuracy. The classifying accuracies for the model on the training set and test set are 92.8% and 89%, respectively. Finally, we use a third-party data to verify the classification model and get a classifying accuracy of 88.8%. The results prove the validity of the classification algorithm proposed in this paper.
This paper proposes an automatic classification algorithm for light curves of eclipsing binary stars based on machine learning. At first the algorithm normalizes the light curves and performs filtering/interpolation to reduce the noise effect in the preprocessing stage, then the Fourier coefficients, which are extracted by FFT from the light curves, are used as the feature vector to train SVM and a classification model is obtained. We implement this algorithm with Python and use the data captured from CALEB/GCVS to validate and discuss the effect of the featured vector, the SVM kernel function and the penalty coefficient on the classifying accuracy. The classifying accuracies for the model on the training set and test set are 92.8% and 89%, respectively. Finally, we use a third-party data to verify the classification model and get a classifying accuracy of 88.8%. The results prove the validity of the classification algorithm proposed in this paper.
引文
[1] 崔辰州, 于策, 肖健, 等. 大数据时代的天文学研究[J]. 科学通报, 2015, 60(增刊1): 445-449.
[2] 张海龙, 聂俊, 赵青, 等. 新疆天文台在线交叉证认服务[J]. 天文研究与技术, 2017, 14(3): 347-355.
[3] LEWIS H, RAFFI G. Advanced software, control, and communication systems for astronomy[C]// Proceedings of the SPIE. 2004, 5496: 65-68.
[4] 卫守林, 刘鹏翔, 王锋, 等. 基于Spark Streaming的明安图射电频谱日像仪实时数据处理[J]. 天文研究与技术, 2017, 14(4): 421-428.
[5] 陈淑鑫, 罗阿理, 孙伟民. R语言应用于LAMOST光谱分析初探[J]. 天文研究与技术, 2017, 14(3): 363-368.
[6] ZHANG Y X, ZHOU X L, ZHAO Y H, et al. Statistical study of 2XMMi-DR3/SDSS-DR8 cross-correlation sample[J]. The Astronomical Journal, 2013, 145(2): 13.
[7] DEVINE T R, GOSEVA-POPSTOJANOVA K, MCLAUGHLIN M. Detection of dispersed radio pulses: a machine learning approach to candidate identification and classification[J]. Monthly Notices of the Royal Astronomical Society, 2016, 459(2): 1519-1532.
[8] 杨远贵, 来春富. 密近双星的Roche势的计算[J]. 淮北师范大学学报 (自然科学版), 2011, 32(1): 29-32.
[9] KIRK B, CONROY K, PR?A A, et al. Kepler eclipsing binary stars. VII. the catalog of eclipsing binaries found in the entire Kepler data-set[J]. The Astronomical Journal, 2016, 151(3): article id. 68(21pp).
[10] POJMANSKI G. The All Sky Automated Survey[J]. Acta Astronomica, 1997, 47: 467-481.
[11] AKERLOF C, AMROSE S, BALSANO R, et al. ROTSE all-sky surveys for variable stars. I. testfields[J]. The Astronomical Journal, 2000, 119(4): article id. 1901(29pp).
[12] POJMANSKI G. The All Sky Automated Survey. catalog of variable stars. I. 0h-6h quarter of the southern hemisphere [J]. Acta Astronomica, 2002, 52: 397-427.
(1)http://caleb.eastern.edu
(2)http://www.sai.msu.su/gcvs/gcvs/intr.htm
(3)https://www.researchgate.net/profile/Y-G_Yang
[2] 张海龙, 聂俊, 赵青, 等. 新疆天文台在线交叉证认服务[J]. 天文研究与技术, 2017, 14(3): 347-355.
[3] LEWIS H, RAFFI G. Advanced software, control, and communication systems for astronomy[C]// Proceedings of the SPIE. 2004, 5496: 65-68.
[4] 卫守林, 刘鹏翔, 王锋, 等. 基于Spark Streaming的明安图射电频谱日像仪实时数据处理[J]. 天文研究与技术, 2017, 14(4): 421-428.
[5] 陈淑鑫, 罗阿理, 孙伟民. R语言应用于LAMOST光谱分析初探[J]. 天文研究与技术, 2017, 14(3): 363-368.
[6] ZHANG Y X, ZHOU X L, ZHAO Y H, et al. Statistical study of 2XMMi-DR3/SDSS-DR8 cross-correlation sample[J]. The Astronomical Journal, 2013, 145(2): 13.
[7] DEVINE T R, GOSEVA-POPSTOJANOVA K, MCLAUGHLIN M. Detection of dispersed radio pulses: a machine learning approach to candidate identification and classification[J]. Monthly Notices of the Royal Astronomical Society, 2016, 459(2): 1519-1532.
[8] 杨远贵, 来春富. 密近双星的Roche势的计算[J]. 淮北师范大学学报 (自然科学版), 2011, 32(1): 29-32.
[9] KIRK B, CONROY K, PR?A A, et al. Kepler eclipsing binary stars. VII. the catalog of eclipsing binaries found in the entire Kepler data-set[J]. The Astronomical Journal, 2016, 151(3): article id. 68(21pp).
[10] POJMANSKI G. The All Sky Automated Survey[J]. Acta Astronomica, 1997, 47: 467-481.
[11] AKERLOF C, AMROSE S, BALSANO R, et al. ROTSE all-sky surveys for variable stars. I. testfields[J]. The Astronomical Journal, 2000, 119(4): article id. 1901(29pp).
[12] POJMANSKI G. The All Sky Automated Survey. catalog of variable stars. I. 0h-6h quarter of the southern hemisphere [J]. Acta Astronomica, 2002, 52: 397-427.
(1)http://caleb.eastern.edu
(2)http://www.sai.msu.su/gcvs/gcvs/intr.htm
(3)https://www.researchgate.net/profile/Y-G_Yang