Loading...
Please wait, while we are loading the content...
Similar Documents
Two data mining algorithms for predicting the condition number of sparse matrices �
| Content Provider | Semantic Scholar |
|---|---|
| Author | Han, Dianwei Zhang, J. B. |
| Copyright Year | 2006 |
| Abstract | We present experimental results of comparing the Modified KNearest Neighbor (M NN) algorithm with Support Vector Machine (SVM) in the predi ction of condition numbers of sparse matrices. Condition num ber of a matrix is an important measure in numerical analysis and linear algebra . However, the direct computation of the condition number of a matrix is very expen sive in terms of CPU and memory cost, and becomes prohibitive for large size m atrices. We use data mining techniques to estimate the condition number of a given sparse matrix. In our previous work, we used Support Vector Machine (S VM) to predict the condition numbers. While SVM is considered a state-of-t he-art classification/regression algorithm, NN is usually used for collaborative filtering tasks. Since prediction can also be interpreted as a classsificatio n/regression task, virtually any supervised learning algorithm (such as NN) can also be applied. Experiments are performed on a publicly available dataset. We con clude that Modified NN (M NN) performs much better than SVM on this particular dataset . |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.math.tamu.edu/~pasciak/ewing60/paper02.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
