eucalyptus

OpenML dataset with id 188

No author found.

Full work available at URL: https://api.openml.org/data/v1/download/3625/eucalyptus.arff

Upload date: 23 April 2014

Dataset Characteristics

Number of classes: 5
Number of features: 20 (numeric: 14, symbolic: 6 and in total binary: 0 )
Number of instances: 736
Number of instances with missing values: 95
Number of missing values: 448

Description

Author: Bruce Bulloch Source: WEKA Dataset Collection - part of the agridatasets archive. This is the true source Please cite: None

Eucalyptus Soil Conservation The objective was to determine which seedlots in a species are best for soil conservation in seasonally dry hill country. Determination is found by measurement of height, diameter by height, survival, and other contributing factors.

It is important to note that eucalypt trial methods changed over time; earlier trials included mostly 15 - 30cm tall seedling grown in peat plots and the later trials have included mostly three replications of eight trees grown. This change may contribute to less significant results.

Experimental data recording procedures which require noting include:

- instances with no data recorded due to experimental recording procedures
  require that the absence of a species from one replicate at a site was
  treated as a missing value, but if absent from two or more replicates at a
  site the species was excluded from the site's analyses.
- missing data for survival, vigour, insect resistance, stem form, crown form
  and utility especially for the data recorded at the Morea Station; this 
  could indicate the death of species in these areas or a lack in collection
  of data.

Attribute Information

 1.  Abbrev - site abbreviation - enumerated
 2.  Rep - site rep - integer
 3.  Locality - site locality in the North Island - enumerated
 4.  Map_Ref - map location in the North Island - enumerated
 5.  Latitude - latitude approximation - enumerated
 6.  Altitude - altitude approximation - integer
 7.  Rainfall - rainfall (mm pa) - integer
 8.  Frosts - frosts (deg. c) - integer
 9.  Year - year of planting - integer
 10. Sp - species code - enumerated
 11. PMCno - seedlot number - integer
 12. DBH - best diameter base height (cm) - real
 13. Ht - height (m) - real
 14. Surv - survival - integer
 15. Vig - vigour - real
 16. Ins_res - insect resistance - real
 17. Stem_Fm - stem form - real
 18. Crown_Fm - crown form - real
 19. Brnch_Fm - branch form - real
 Class:
 20. Utility - utility rating - enumerated

Relevant papers

Bulluch B. T., (1992) Eucalyptus Species Selection for Soil Conservation in Seasonally Dry Hill Country - Twelfth Year Assessment New Zealand Journal of Forestry Science 21(1): 10 - 31 (1991)

Kirsten Thomson and Robert J. McQueen (1996) Machine Learning Applied to Fourteen Agricultural Datasets. University of Waikato Research Report https://www.cs.waikato.ac.nz/ml/publications/1996/Thomson-McQueen-96.pdf + the original publication:

This page was built for dataset: eucalyptus