| Creator |
|
|
|
|
|
|
|
|
|
| Language |
|
| Publisher |
|
|
|
| Date |
|
| Source Title |
|
| Vol |
|
| Publication Type |
|
| Access Rights |
|
| Related DOI |
|
| Related DOI |
|
|
|
| Related URI |
|
|
|
| Relation |
|
|
|
|
|
| Abstract |
In this paper, we develop the method for estimating the correlation dimension proposed by Kawaguchi (2002) and apply it to geomagnetic pulsations. It is found that the estimated values of the correlat...ion dimension tend to drop at auroral breakups. This is developed into the detection of auroral breakups. At first, using the data that is easy to detect visually auroral breakups by the waveform, the training data is expressed with the pair of the estimates of the correlation dimension and the label which codes as +1 for auroral breakups and −1 for non-auroral breakups. Next, we construct the discriminant function by the support vector machine using the training data. Using the training data, the error for the discrimination is computed. The error is found to be 11.3%. This suggests the usefulness of the estimates of the correlation dimension in the discrimination.show more
|
Export Mendeley
2004-5