Data

cjs

in_preparation
ARFF
Publicly available Visibility: public Uploaded 28-09-2014 by Joaquin Vanschoren

0 likes downloaded by 2 people , 2 total downloads 0 issues 0 downvotes

0 likes downloaded by 2 people , 2 total downloads 0 issues 0 downvotes

Issue | #Downvotes for this reason | By |
---|

Loading wiki

Help us complete this description
Edit

Author:
Source: Unknown - Date unknown
Please cite:
---------------------------------------------------------------------------
Short description
---------------------------------------------------------------------------
Data on tree growth used in the Case Study published in the September, 1995
issue of the Canadian Journal of Statistics
---------------------------------------------------------------------------
Permission
---------------------------------------------------------------------------
This data set was been provided by Dr. Fernando Camacho,
Ontario Hydro Technologies, 800 Kipling Ave, Toronto Canada M3Z 5S4.
It forms the basis of the Case Study in Data Analysis published in
the Canadian Journal of Statistics, September 1995.
It can be freely used for non-commercial purposes, as long as proper
acknowledgement to the source and to the Canadian Journal of Statistics
is made.
---------------------------------------------------------------------------
Description
---------------------------------------------------------------------------
The effects of the Growth Regulators Paclobutrazol (PP 333)
and Flurprimidol (EL-500) on the Number and Length of Internodes
in Terminal Sprouts Formed on Trimmed Silver Maple Trees.
Introduction:
The trimming of trees under distribution lines on city streets and
in rural areas is a major problem and expense for electrical
utilities. Such operations are routinely performed at intervals of
one to eight years depending upon the individual species growth rate
and the amount of clearance required. Ontario Hydro trims about
500,000 trees per year at a cost of about $25 per tree.
Much effort has been spent in developing chemicals for the horticultural
industry to retard the growth of woody and herbaceous plants. Recently,
a group of new growth regulators was introduced which was shown to be
effective in controlling the growth of trees without producing
noticeable injury symptoms. In this group are PP 333 ( common name
paclobutrazol) (2RS, 3RS - 1 -(4-chlorophenyl) - 4,4 - dimethyl - 2 -
(1,2,4-triazol-l-yl) pentan - 3- ol and EL-500 (common name flurprimidol
and composition alpha - (1-methylethyl) - alpha - [4-(trifluromethoxyl)
phenyl] - 5- pyrimidine - methanol). Both EL-500 and PP-333 have been
reported to control excessive sprout growth in a number of species
when applied as a foliar spray, as a soil drench, or by trunk injection.
Sprout length is a function of both the number of internodes and
the length of the individual internodes in the sprout. While there
have been many reports that both PP 333 and EL-500 cause a reduction
in the length of internodes formed in sprouts on woody plants treated
with the growth regulators, there has been but one report that EL-500
application to apple trees resulted in a reduction of the number
of internodes formed per sprout.
The purpose of the present study was to investigate the length of the
terminal sprouts, the length of the individual internodes in those
sprouts, and the number of internodes in trimmed silver maple trees
following trunk injection with the growth regulators PP 333 and EL-500.
Experimental Details.
Multistemmed 12-year-old silver maple trees growing at Wesleyville,
Ontario were trunk injected with methanolic solutions of EL-500
and PP-333 in May of 1985 using a third generation Asplundh
tree injector.
Two different application rates (20 g/L and 4 g/L) were used for each
chemical. The volume of solution (and hence the amount of active
ingredient) injected into each tree was determined from the diameter
of the tree, using the formula: vol(mL) = (dbh)*(dbh)*.492 where dbh
is the diameter at breast height. Two sets of control trees were
included in the experiment. In one set, tree received no injection
(control) and in a second set, the trees were injected with
methanol, the carrier in the growth regulator solutions. Ten trees,
chosen at random, were used in each of the control and experimental
sets. Prior to injection, all the trees were trimmed by a forestry
crew, with their heights being reduced by about one third.
In January 1987, twenty months after the trees were injected, between
six and eight limbs were removed at random from the bottom two-thirds
of the canopy of each of the ten trees in each experimental and control
set. The limbs were returned to the laboratory and the length of all
the terminal sprouts, the lengths of the individual internodes, and
the number of internodes recorded. Between one and 25 terminal
sprouts were found on each limb collected. Sprouts which had a
length of 1 cm or less were recorded as being 1 cm in length.
In such spouts, the internode lengths were not measured, but were
calculated from the total length of the sprout and the number
of internodes counted. Internode lengths were then expressed to one
decimal place. In two instances, one of the ten trees in a set
could not be sampled because limb removal would have jeopardized the
health of the tree over the long-term.
Data set:
Each of the records represents a terminal sprout and contains the
following information:
N - the sprout number
TR - treatment 1 - control
2 - methanol control
3 - PP 333 20g/L
4 - PP 333 4g/L
5 - EL 500 20g/L
6 - EL 500 4g/L
TREE - tree id
BR - branch id
TL - total sprout length (cm)
IN - number of internodes on the sprout
INTER- a list of the lengths of the internodes in the sprout,
starting from the base of the sprout (1-29 entries)
Sprouts 1868 to 1879 do not have branch identification data.
Here is a portion of the data file.
1 1 G28 A 75.0 15 1.0 2.3 7.4 8.6 6.7 7.2 6.6 6.2 5.5 5.0 5.4 5.0 4.4 2.6 0.8
2 1 G28 B 18.0 7 0.7 1.3 4.0 5.2 2.8 2.2 1.5
3 1 G28 C 46.0 11 0.5 1.0 4.3 8.8 6.8 7.6 6.2 4.7 3.5 2.2 0.5
4 1 G28 C 16.0 8 0.5 1.2 3.5 4.2 2.2 2.8 1.5 0.1
5 1 G28 D 56.0 16 0.5 1.0 3.5 8.2 5.3 6.2 4.8 4.7 3.1 3.8 3.3 4.3 3.5 2.4 1.2 0.3
.
.
.
10 1 G28 G 1.0 3 0.3 0.5 0.3
11 1 G28 G 1.0 3 0.2 0.6 0.5
12 1 G28 G 4.0 5 0.5 0.7 1.3 1.0 0.1
13 1 G28 G 16.0 8 0.7 1.2 3.2 4.7 2.8 1.8 1.0 0.2
14 1 G28 G 3.0 4 0.5 1.0 1.2 0.1
15 1 G28 G 21.0 10 0.5 0.8 2.4 4.0 3.0 3.2 2.6 2.1 1.4 0.7
.
.
.
18 1 M33 A 9.0 5 1.2 2.2 3.1 2.2 0.7
19 1 M33 A 20.5 7 0.6 3.0 5.8 5.4 3.6 1.7 0.4
20 1 M33 A 47.0 13 0.5 2.4 2.8 3.5 3.1 3.6 3.8 5.1 8.6 6.5 5.6 1.1 0.3
21 1 M33 A 4.0 4 0.4 2.0 1.7 0.2
22 1 M33 A 5.0 5 0.2 1.2 1.6 1.5 0.7
.
.
.
---------------------------------------------------------------------------
Information about the dataset
CLASSTYPE: nominal
CLASSINDEX: 2

TR (target) | nominal | 6 unique values 0 missing | |

N | numeric | 2796 unique values 0 missing | |

TREE | nominal | 57 unique values 0 missing | |

BR | nominal | 10 unique values 12 missing | |

TL | numeric | 170 unique values 0 missing | |

IN | numeric | 26 unique values 0 missing | |

INTERNODE_1 | numeric | 30 unique values 0 missing | |

INTERNODE_2 | numeric | 68 unique values 64 missing | |

INTERNODE_3 | numeric | 117 unique values 637 missing | |

INTERNODE_4 | numeric | 128 unique values 1634 missing | |

INTERNODE_5 | numeric | 113 unique values 1999 missing | |

INTERNODE_6 | numeric | 115 unique values 2172 missing | |

INTERNODE_7 | numeric | 99 unique values 2308 missing | |

INTERNODE_8 | numeric | 95 unique values 2406 missing | |

INTERNODE_9 | numeric | 86 unique values 2489 missing | |

INTERNODE_10 | numeric | 86 unique values 2543 missing | |

INTERNODE_11 | numeric | 85 unique values 2578 missing | |

INTERNODE_12 | numeric | 78 unique values 2608 missing | |

INTERNODE_13 | numeric | 76 unique values 2634 missing | |

INTERNODE_14 | numeric | 72 unique values 2653 missing | |

INTERNODE_15 | numeric | 66 unique values 2660 missing | |

INTERNODE_16 | numeric | 52 unique values 2678 missing | |

INTERNODE_17 | numeric | 51 unique values 2711 missing | |

INTERNODE_18 | numeric | 43 unique values 2725 missing | |

INTERNODE_19 | numeric | 35 unique values 2740 missing | |

INTERNODE_20 | numeric | 23 unique values 2753 missing | |

INTERNODE_21 | numeric | 18 unique values 2770 missing | |

INTERNODE_22 | numeric | 10 unique values 2779 missing | |

INTERNODE_23 | numeric | 9 unique values 2786 missing | |

INTERNODE_24 | numeric | 7 unique values 2789 missing | |

INTERNODE_25 | numeric | 3 unique values 2792 missing | |

INTERNODE_26 | numeric | 1 unique values 2795 missing | |

INTERNODE_27 | numeric | 1 unique values 2795 missing | |

INTERNODE_28 | numeric | 1 unique values 2795 missing | |

INTERNODE_29 | numeric | 1 unique values 2795 missing |

2.33

An estimate of the amount of irrelevant information in the attributes regarding the class. Equals (MeanAttributeEntropy - MeanMutualInformation) divided by MeanMutualInformation.

0.05

First quartile of mutual information between the nominal attributes and the target attribute.

1

Kappa coefficient achieved by the landmarker weka.classifiers.bayes.NaiveBayes -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W

0.93

Kappa coefficient achieved by the landmarker weka.classifiers.trees.RandomTree -depth 3

1

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.J48 -C .001

24.33

Average number of distinct values among the attributes of the nominal type.

1

Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 1

1

Area Under the ROC Curve achieved by the landmarker weka.classifiers.lazy.IBk -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W

28.36

Standard deviation of the number of distinct values among attributes of the nominal type.

0.93

First quartile of standard deviation of attributes of the numeric type.

1

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.REPTree -L 2

0

Error rate achieved by the landmarker weka.classifiers.lazy.IBk -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W

1

Kappa coefficient achieved by the landmarker weka.classifiers.lazy.IBk -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W

-0.25

Second quartile (Median) of kurtosis among attributes of the numeric type.

1

Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 2

3.01

Second quartile (Median) of means among attributes of the numeric type.

1

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.REPTree -L 3

0.81

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.DecisionStump

1.29

Second quartile (Median) of mutual information between the nominal attributes and the target attribute.

0.55

Error rate achieved by the landmarker weka.classifiers.trees.DecisionStump

0.05

Minimal mutual information between the nominal attributes and the target attribute.

0.72

Second quartile (Median) of skewness among attributes of the numeric type.

1

Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 3

0.29

Kappa coefficient achieved by the landmarker weka.classifiers.trees.DecisionStump

6

The minimal number of distinct values among attributes of the nominal type.

2.35

Second quartile (Median) of standard deviation of attributes of the numeric type.

0.98

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.RandomTree -depth 1

2.52

Maximum mutual information between the nominal attributes and the target attribute.

0.06

Error rate achieved by the landmarker weka.classifiers.trees.RandomTree -depth 1

1.96

Number of attributes needed to optimally describe the class (under the assumption of independence among attributes). Equals ClassEntropy divided by MeanMutualInformation.

57

The maximum number of distinct values among attributes of the nominal type.

3.16

Third quartile of kurtosis among attributes of the numeric type.

0.93

Kappa coefficient achieved by the landmarker weka.classifiers.trees.RandomTree -depth 1

1

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.J48 -C .00001

1

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.DecisionStump -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W

0.98

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.RandomTree -depth 2

2.52

Third quartile of mutual information between the nominal attributes and the target attribute.

0

Error rate achieved by the landmarker weka.classifiers.trees.DecisionStump -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W

0.06

Error rate achieved by the landmarker weka.classifiers.trees.RandomTree -depth 2

0.98

Area Under the ROC Curve achieved by the landmarker weka.classifiers.bayes.NaiveBayes

1.79

Third quartile of skewness among attributes of the numeric type.

1

Kappa coefficient achieved by the landmarker weka.classifiers.trees.DecisionStump -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W

0.93

Kappa coefficient achieved by the landmarker weka.classifiers.trees.RandomTree -depth 2

1

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.J48 -C .0001

-0.55

First quartile of kurtosis among attributes of the numeric type.

2.91

Third quartile of standard deviation of attributes of the numeric type.

1

Area Under the ROC Curve achieved by the landmarker weka.classifiers.bayes.NaiveBayes -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W

0.98

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.RandomTree -depth 3

1.29

Average mutual information between the nominal attributes and the target attribute.

1

Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.REPTree -L 1

0

Error rate achieved by the landmarker weka.classifiers.bayes.NaiveBayes -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W

0.06

Error rate achieved by the landmarker weka.classifiers.trees.RandomTree -depth 3