9 | | The following datasets are used in experiments in the thesis: |
10 | | * Artificial benchmark datasets |
11 | | * Friedman-I |
12 | | * Friedman-II |
13 | | * Breiman-I |
14 | | * Real-world datasets |
15 | | * Chemical-I |
16 | | * Chemical-II |
17 | | * Financial-I |
18 | | * Macro-Economic |
19 | | * Housing |
| 9 | The following datasets are used in experiments in the thesis. |
| 10 | ==== Artificial benchmark datasets ==== |
| 11 | ===== Friedman-I ===== |
| 12 | This dataset is described in \cite{friedman1991} where it is used to benchmark the multi-variate adaptive regression splines (MARS) algorithm. The signal-to-noise ratio in this dataset is rather low so it is difficult to rediscover the generating function especially the terms below the noise level x4 and x5). |
| 13 | |
| 14 | [[Image(attachment:friedman-I.png)]] |
| 15 | |
| 16 | Variables x01,..., x10 are sampled uniformly from the unit hypercube. |
| 17 | Epsilon is generated from the standard normal distribution. |
| 18 | |
| 19 | [export:/trunk/misc/pub/2010/theses/gkronber/friedman-I.csv friedman-I.csv] |
| 20 | |
| 21 | ===== Friedman-II ===== |
| 22 | [export:/trunk/misc/pub/2010/theses/gkronber/friedman-II.csv friedman-II.csv] |
| 23 | |
| 24 | ===== Breiman-I ===== |
| 25 | [export:/trunk/misc/pub/2010/theses/gkronber/breiman-I.csv breiman-I.csv] |
| 26 | |
| 27 | ==== Real-world datasets ==== |
| 28 | |
| 29 | ===== Chemical-I ===== |
| 30 | [export:/trunk/misc/pub/2010/theses/gkronber/chemical-I.csv chemical-I.csv] |
| 31 | |
| 32 | ===== Chemical-II ===== |
| 33 | [export:/trunk/misc/pub/2010/theses/gkronber/chemical-II.csv chemical-II.csv] |
| 34 | |
| 35 | ===== Financial-I ===== |
| 36 | [export:/trunk/misc/pub/2010/theses/gkronber/financial-I.csv financial-I.csv] |
| 37 | |
| 38 | ===== Macro-Economic ===== |
| 39 | [export:/trunk/misc/pub/2010/theses/gkronber/macro-economic.csv macro-economic.csv] |
| 40 | |
| 41 | ===== Housing ===== |
| 42 | [export:/trunk/misc/pub/2010/theses/gkronber/housing.csv housing.csv] |
| 43 | |