Support vector machines and kd-tree for separating quasars from large survey data bases
Support vector machines and kd-tree for separating quasars from large survey data bases
复制标题
用于从大型调查数据库中分离类星体的支持向量机和 kd 树
DOI：
10.1111/j.1365-2966.2008.13070.x
复制
发表时间：
2008-02
影响因子：
4.8
通讯作者：
高丹
中科院分区：
文献类型：
--
作者：
赵永恒;张彦霞;高丹
我们比较了k维树（kd-tree）和支持向量机（svm）两种自动分类算法在斯隆数字巡天（SDSS）和两微米全巡天（2MASS）目录数据库中将类星体与恒星分离的性能。这两种算法是在SDSS和2MASS物体的子集上训练的，这些物体的性质是通过光谱已知的。我们选择不同的属性组合作为输入模式，只使用光度数据训练分类器，并给出两种方法的分类结果。计算精度和召回率、真阳性率和真阴性率、F-measure、G-mean和加权精度等性能指标来评估两种算法的性能。研究表明，kd-tree和svm都是有效的点源自动分类算法。支持向量机的准确率略高，但kd-tree所需的计算时间更少。给定基于不同参数（例如大小，颜色信息）的不同输入模式，我们得出结论，kd-tree和svm在特征较少的情况下表现出更好的性能。此外，我们的结果还表明，使用四种颜色（u - g, g - r， r - i和- z）和基于SDSS模型星等的r星等的精度加起来是最高的。利用kd-tree和svm训练的分类器可以解决虚拟天文台（VO）面临的自动分类问题；此外，它们都可以应用于大型巡天项目中类星体候选者的光度预选，以优化望远镜的效率。
We compare the performance of two automated classification algorithms, k-dimensional tree (kd-tree) and support vector machines (SVMs), to separate quasars from stars in the data bases of the Sloan Digital Sky Survey (SDSS) and the Two-Micron All Sky Survey (2MASS) catalogues. The two algorithms are trained on subsets of SDSS and 2MASS objects whose nature is known via spectroscopy. We choose different attribute combination as input patterns to train the classifier using photometric data only and present the classification results obtained by these two methods. Performance metrics, such as precision and recall, true positive rate and true negative rate, F-measure, G-mean and Weighted Accuracy, are computed to evaluate the performance of the two algorithms. The study shows that both kd-tree and SVMs are effective automated algorithms to classify point sources. SVMs show slightly higher accuracy, but kd-tree requires less computation time. Given different input patterns based on various parameters (e.g. magnitudes, colour information), we conclude that both kd-tree and SVMs show better performance with fewer features. What is more, our results also indicate that the accuracy using the four colours (u - g, g - r, r - i andi - z) and r magnitude based on SDSS model magnitudes adds up to the highest value. The classifiers trained by kd-tree and SVMs can be used to solve the automated classification problems faced by the virtual observatory (VO); moreover, they can all be applied for the photometric preselection of quasar candidates for large survey projects in order to optimise the efficiency of telescopes.
登录
查看更多内容
DOI：
10.1086/507440
发表时间：
2006-06
期刊：
The Astrophysical Journal
影响因子：
--
作者：
N. Ball;R. Brunner;A. Myers;D. Tcheng
通讯作者：
N. Ball;R. Brunner;A. Myers;D. Tcheng
影响因子：
2.8
作者：
M. Qu;F. Shih;J. Jing;Haimin Wang
通讯作者：
M. Qu;F. Shih;J. Jing;Haimin Wang
DOI：
10.1111/j.1365-2966.2007.12129.x
发表时间：
2007-06
影响因子：
4.8
作者：
D. Wang;Y. X. Zhang;C. Liu;Y. H.Zhao
通讯作者：
D. Wang;Y. X. Zhang;C. Liu;Y. H.Zhao
DOI：
--
发表时间：
2003
期刊：
Comput. Linguistics
影响因子：
--
作者：
Roberto Basili
通讯作者：
Roberto Basili
DOI：
10.1086/300839
发表时间：
1999-01
期刊：
The Astronomical Journal
影响因子：
--
作者：
A. Connolly;A. Connolly;A. Szalay
通讯作者：
A. Connolly;A. Connolly;A. Szalay