In this study we apply Support Vector Machine (SVM) to the prediction of geo-effective halo coronal mass ejections (CMEs). The SVM, which is one of machine learning algorithms, is used for the purpose of classification and regression analysis. We use halo and partial halo CMEs from January 1996 to April 2010 in the SOHO/LASCO CME Catalog for training and prediction. And we also use their associated X-ray flare classes to identify front-side halo CMEs (stronger than B1 class), and the Dst index to determine geo-effective halo CMEs (stronger than -50 nT). The combinations of the speed and the angular width of CMEs, and their associated X-ray classes are used for input features of the SVM. We make an attempt to find the best model by using cross-validation which is processed by changing kernel functions of the SVM and their parameters. As a result we obtain statistical parameters for the best model by using the speed of CME and its associated X-ray flare class as input features of the SVM: Accuracy=0.66, PODy=0.76, PODn=0.49, FAR=0.72, Bias=1.06, CSI=0.59, TSS=0.25. The performance of the statistical parameters by applying the SVM is much better than those from the simple classifications based on constant classifiers.
We have examined the relationships among coronal holes (CHs), corotating interaction regions (CIRs), and geomagnetic storms in the period 1996 -2003. We have identified 123 CIRs with forward and reverse shock or wave features in ACE and Wind data and have linked them to coronal holes shown in National Solar Observatory/Kitt Peak (NSO/KP) daily He I 10 830 Å maps considering the Sun -Earth transit time of the solar wind with the observed wind speed. A sample of 107 CH -CIR pairs is thus identified. We have examined the magnetic polarity, location, and area of the CHs as well as their association with geomagnetic storms (Dst ≤ −50 nT). For all pairs, the magnetic polarity of the CHs is found to be consistent with the sunward (or earthward) direction of the interplanetary magnetic fields (IMFs), which confirms the linkage between the CHs and the CIRs in the sample. Our statistical analysis shows that (1) the mean longitude of the center of CHs is about 8°E, (2) 74% of the CHs are located between 30°S and 30°N (i.e., mostly in the equatorial regions), (3) 46% of the CIRs are associated with geomagnetic storms, (4) the area of geoeffective coronal holes is found to be larger than 0.12% of the solar hemisphere area, and (5) the maximum convective electric field E y in the solar wind is much more highly correlated with the Dst index than any other solar or interplanetary parameter. In addition, we found that there is also a semiannual variation of CIR-associated geomagnetic storms and discovered new tendencies as follows: For negative-polarity coronal holes, the percentage (59%; 16 out of 27 events) of CIRs associated with geomagnetic storms in the first half of the year is much larger than that (25%; 6 out of 24 events) in the second half of the year and the occurrence percentage (63%; 15 out of 24 events) of CIR-associated storms in the southern hemisphere is significantly larger than that (26%; 7 out of 27 events) in the northern hemisphere. Positive-polarity coronal holes exhibit an opposite tendency.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.