Summary

Artificial data generated from the Friedman function, part of a collection of 80 data sets. This particular set has 5 features, 1 output (6th feature), colinearity degree 2, and 250 instances.

License

unknown (from Weka repository)

Dependencies

Tags

arff colinearity Friedman-function slurped Weka

Attribute Types

Floating Point

Download

# Instances: 250 / # Attributes: 6
HDF5 (22.3 KB) XML CSV ARFF LibSVM Matlab Octave

Original Data Format

arff

Name

fri_c2_250_5

Version mldata

0

Comment

Names

oz1,oz2,oz3,oz4,oz5,oz6,

Types

1. numeric
2. numeric
3. numeric
4. numeric
5. numeric
6. numeric

Data (first 10 data points)

oz1	oz2	oz3	oz4	oz5	oz6
0.74...	0.53...	0.35...	-0.57...	0.52...	-1.69...
-1.04...	-0.87...	-0.54...	-1.57...	-0.84...	-0.00...
-0.54...	-0.52...	-0.94...	-0.80...	0.79...	1.26...
2.13...	1.52...	-0.27...	-0.15...	0.84...	-0.27...
1.26...	1.44...	-0.07...	-0.48...	-0.61...	0.50...
-1.64...	-1.49...	-0.48...	1.64...	1.10...	0.64...
-1.86...	-1.68...	-1.11...	1.53...	-1.69...	0.53...
0.48...	-0.08...	0.28...	1.12...	1.61...	0.16...
-1.96...	-1.56...	-1.07...	-1.35...	-0.92...	-0.28...
-0.52...	-0.08...	-0.16...	-0.15...	-0.24...	0.91...
...	...	...	...	...	...

Description

A zip file containing 80 artificial datasets generated from the Friedman function donated by Dr. M. Fatih Amasyali (Yildiz Technical Unversity) (Friedman-datasets.zip, 5,802,204 Bytes)

Information adapted from original readme:

The Friedman_datasets.zip folder contains 80 artificially generated Friedman (*) datasets. The dataset names are coded as "fri_colinearintydegree_samplenumber_featurenumber".

Friedman is the one of the most used function for data generation (Friedman, 1999). Friedman function includes both linear and non-linear relations between input and output. And a normalized noise (e) is added to the output. The Friedman function is given at below.

y = 10 * sin(pi * x1 * x2) + 20 * (x3 - 0.5)^2 = 10 * X4 + 5 * X5 + e

In original Friedman function, there are 5 features for input. In our experiments, to measure the effects of non-related features, the additional features are added into the datasets. The added features are independent from output.

However, to measure the algorithm's robustness to the colinearity, the datasets are generated with 5 different colinearity degrees. The colinearity degrees is the number of features depend on other features.

The generated Friedman dataset's parameters and values are given below: The number of features: 5 10 25 50 100 (only the first 5 one is related to the output. The rest are completely random) The number of samples: 100 250 500 1000 Colinearity degrees: 0 1 2 3 4 For the datasets with colinearity degree 4, the numbers of features are generated as 10, 25, 50 and 100. The other datasets have 5, 10, 25 and 50 features.

As a result, 80 artificial datasets are generated by (4 different feature number * 4 different sample number * 5 different colinearity degree)

The last attribute in each file is the target.

*J.H. Friedman (1999). Stochastic Gradient Boosting

URLs

(No information yet)

Publications

Data Source

http://www.ce.yildiz.edu.tr/en/myindex.php?id=14 http://www.ce.yildiz.edu.tr/

Measurement Details

Usage Scenario

revision 1

by mldata on 2010-11-06 10:02

revision 2

by jaakkopeltonen on 2010-11-22 18:13

revision 3

by jaakkopeltonen on 2011-09-14 14:55

No one has posted any comments yet. Perhaps you would like to be the first?

Leave a comment

This item was downloaded 8448 times and viewed 8054 times.

No Tasks yet on dataset Friedman-datasets fri_c2_250_5