Evaluating the Optical Classification of Fermi BCUs Using Machine Learning

Evaluating the Optical Classification of Fermi BCUs Using Machine Learning
复制标题

使用机器学习评估费米 BCU 的光学分类

DOI:
10.3847/1538-4357/ab0383
复制
发表时间:
2019-02
期刊:
ApJ
影响因子:
--
通讯作者:
Yue Yin
Yue Yin
中科院分区:
其他
文献类型:
--
作者:
Shi-Ju Kang;Jun-Hui Fan;Weiming Mao;Qingwen Wu;Jianchao Feng;Yue Yin

文献摘要

参考文献

被引文献

相似文献

在Fermi-LAT(3LAC)Clean Sample探测到的第三个活动星系核星表中,有402个不确定类型的耀变体(BCU)。由于天文观测条件的限制或自身性质的限制,耀变体的光谱分类比较困难。使用机器学习算法对BCU进行潜在分类至关重要。基于3LAC清洁样品,我们收集了1420个费米耀变体,包括γ射线光子谱指数、射电通量、通量密度、曲线显著性、100-300 MeV、0.3-1 GeV和10-100 GeV的积分光子通量和变率指数等8个参数。在这里,我们应用四种不同的监督机器学习(SML)算法(决策树,随机森林,支持向量机和MPEG-Gaussian有限混合模型)来评估基于直接观测特性的BCU分类。所有四种方法都可以表现得非常好,更准确,可以有效地预测费米BCU的分类。评价结果表明,这些方法(SML)的结果是有效的和稳健的,在400个BCU中,约1/4的源是平谱射电类星体(FSRQ),3/4的源是BL Lacertae(BL拉克),这与最近的一些结果是一致的。虽然SML的准确性受到许多因素的影响,但在FSRQ和BL拉克之间的固定比例为1:3时,结果是稳定的,这表明SML可以提供一种有效的方法来评估潜在的BCU分类。在这四种方法中,高斯混合模型对我们的训练样本具有最高的准确性(4/5,种子= 123)。
In the third catalog of active galactic nuclei detected by the Fermi-LAT (3LAC) Clean Sample, there are 402 blazar candidates of uncertain type (BCUs). Due to the limitations of astronomical observation or intrinsic properties, it is difficult to classify blazars using optical spectroscopy. The potential classification of BCUs using machine-learning algorithms is essential. Based on the 3LAC Clean Sample, we collect 1420 Fermi blazars with eight parameters of γ-ray photon spectral index; radio flux; flux density; curve significance; the integral photon flux in 100–300 MeV, 0.3–1 GeV, and 10–100 GeV; and variability index. Here we apply four different supervised machine-learning (SML) algorithms (decision trees, random forests, support vector machines, and Mclust Gaussian finite mixture models) to evaluate the classification of BCUs based on the direct observational properties. All four methods can perform exceedingly well with more accuracy and can effectively forecast the classification of Fermi BCUs. The evaluating results show that the results of these methods (SML) are valid and robust, where about one-fourth of sources are flat-spectrum radio quasars (FSRQs) and three-fourths are BL Lacertae (BL Lacs) in 400 BCUs, which are consistent with some other recent results. Although a number of factors influence the accuracy of SML, the results are stable at a fixed ratio 1:3 between FSRQs and BL Lacs, which suggests that the SML can provide an effective method to evaluate the potential classification of BCUs. Among the four methods, Mclust Gaussian Mixture Modeling has the highest accuracy for our training sample (4/5, seed = 123).
DOI: 10.3390/galaxies4040036
发表时间: 2016-09
期刊: --
影响因子: --
作者:
G. Ghisellini
通讯作者: G. Ghisellini
DOI: 10.3847/1538-4357/aa6005
发表时间: 2017-02
期刊: The Astrophysical Journal
影响因子: --
作者:
Shi-Ju Kang
通讯作者: Shi-Ju Kang
DOI: 10.1088/0004-637x/743/2/171
发表时间: 2011-08
期刊: --
影响因子: --
作者:
The Fermi-LAT Collaboration
通讯作者: The Fermi-LAT Collaboration
DOI: 10.1093/mnras/stw1830
发表时间: 2016-07
影响因子: 4.8
作者:
G. Chiaro;D. Salvetti;G. L. Mura;M. Giroletti;D. Thompson;D. Bastieri
通讯作者: G. Chiaro;D. Salvetti;G. L. Mura;M. Giroletti;D. Thompson;D. Bastieri
DOI: 10.1093/mnras/stx1328
发表时间: 2017-05
影响因子: 4.8
作者:
D. Salvetti;G. Chiaro;G. L. Mura;D. Thompson
通讯作者: D. Salvetti;G. Chiaro;G. L. Mura;D. Thompson