Data

pbc

active
ARFF
Publicly available Visibility: public Uploaded 23-04-2014 by Jan van Rijn

0 likes downloaded by 1 people , 1 total downloads 0 issues 0 downvotes

0 likes downloaded by 1 people , 1 total downloads 0 issues 0 downvotes

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

Loading wiki

Help us complete this description
Edit

Author:
Source: Unknown -
Please cite:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Case number deleted. X treated as the class attribute.
As used by Kilpatrick, D. & Cameron-Jones, M. (1998). Numeric prediction
using instance-based learning with encoding length selection. In Progress
in Connectionist-Based Information Systems. Singapore: Springer-Verlag.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
NAME: PBC Data
SIZE: 418 observations, 20 variables
DESCRIPTIVE ABSTRACT:
Below is a description of the variables recorded from the Mayo Clinic trial
in primary biliary cirrhosis (PBC) of the liver conducted between 1974 and
1984. A total of 424 PBC patients, referred to Mayo Clinic during
that ten-year interval, met eligibility criteria for the randomized placebo
controlled trial of the drug D-penicillamine. The first 312 cases in the data
set participated in the randomized trial, and contain largely complete data.
The additional 112 cases did not participate in the clinical trial, but
consented to have basic measurements recorded and to be followed for survival.
Six of those cases were lost to follow-up shortly after diagnosis, so there
are data here on an additional 106 cases as well as the 312 randomized
participants. Missing data items are denoted by ".". At least one space
separates each variable in the .DAT file. Censoring was due to liver
transplantation for twenty-five subjects with the following case numbers:
5, 105, 111, 120, 125, 158, 183, 241, 246, 247, 254, 263, 264, 265, 274,
288, 291, 295, 297, 345, 361, 362, 375, 380, 383.
SOURCE: Counting Processes and Survival Analysis by T. Fleming &
D. Harrington, (1991), published by John Wiley & Sons.
VARIABLE DESCRIPTIONS:
The data are in free format. That is, at least one blank space separates
each variable. The variables contained in the .DAT are:
N: Case number.
X: The number of days between registration and the earlier of
death, liver transplantation, or study analysis time in July, 1986.
D: 1 if X is time to death, 0 if time to censoring
Z1: Treatment Code, 1 = D-penicillamine, 2 = placebo.
Z2: Age in years. For the first 312 cases, age was calculated by
dividing the number of days between birth and study registration by 365.
Z3: Sex, 0 = male, 1 = female.
Z4: Presence of ascites, 0 = no, 1 = yes.
Z5: Presence of hepatomegaly, 0 = no, 1 = yes.
Z6: Presence of spiders 0 = no, 1 = Yes.
Z7: Presence of edema, 0 = no edema and no diuretic therapy for
edema; 0.5 = edema present for which no diuretic therapy was given, or
edema resolved with diuretic therapy; 1 = edema despite diuretic therapy
Z8: Serum bilirubin, in mg/dl.
Z9: Serum cholesterol, in mg/dl.
Z10: Albumin, in gm/dl.
Z11: Urine copper, in mg/day.
Z12: Alkaline phosphatase, in U/liter.
Z13: SGOT, in U/ml.
Z14: Triglycerides, in mg/dl.
Z15: Platelet count; coded value is number of platelets
per-cubic-milliliter of blood divided by 1000.
Z16: Prothrombin time, in seconds.
Z17: Histologic stage of disease, graded 1, 2, 3, or 4.
STORY BEHIND THE DATA:
Between January, 1974 and May, 1984, the Mayo Clinic conducted a
double-blinded randomized trial in primary biliary cirrhosis of the liver
(PBC), comparing the drug D-penicillamine (DPCA) with a placebo. There
were 424 patients who met the eligibility criteria seen at the Clinic while
the trial was open for patient registration. Both the treating physician and
the patient agreed to participate in the randomized trial in 312 of the 424
cases. The date of randomization and a large number of clinical, biochemical,
serologic, and histologic parameters were recorded for each of the 312
clinical trial patients. The data from the trial were analyzed in 1986 for
presentation in the clinical literature. For that analysis, disease and
survival status as of July, 1986, were recorded for as many patients as
possible. By that date, 125 of the 312 patients had died, with only 11
not attributable to PBC. Eight patients had been lost to follow up, and 19
had undergone liver transplantation.
PBC is a rare but fatal chronic liver disease of unknown cause,
with a prevalence of about 50-cases-per-million population. The primary
pathologic event appears to be the destruction of interlobular bile ducts,
which may be mediated by immunologic mechanisms. The data discussed here are
important in two respects. First, controlled clinical trials are difficult to
complete in rare diseases, and this case series of patients uniformly
diagnosed, treated, and followed is the largest existing for PBC. The
treatment comparison in this trial is more precise than in similar trials
having fewer participants and avoids the bias that may arise in comparing
a case series to historical controls. Second, the data present an
opportunity to study the natural history of the disease. We will see that,
despite the immunosuppressive properties of DPCA, there are no detectable
differences between the distributions of survival times for the DPCA and
placebo treatment groups. This suggests that these groups can be combined
in studying the association between survival time from randomization and
clinical and other measurements. In the early to mid 1980s, the rate of
successful liver transplant increased substantially, and transplant has
become an effective therapy for PBC. The Mayo Clinic data set is therefore
one of the last allowing a study of the natural history of PBC in patients
who were treated with only supportive care or its equivalent. The PBC data
can be used to: estimate a survival distribution; test for differences
between two groups; and estimate covariate effects via a regression
model.

class (target) | numeric | 399 unique values 0 missing | |

D | nominal | 2 unique values 0 missing | |

Z1 | nominal | 2 unique values 106 missing | |

Z2 | numeric | 345 unique values 0 missing | |

Z3 | nominal | 2 unique values 106 missing | |

Z4 | nominal | 2 unique values 106 missing | |

Z5 | nominal | 2 unique values 106 missing | |

Z6 | nominal | 2 unique values 106 missing | |

Z7 | nominal | 3 unique values 0 missing | |

Z8 | numeric | 98 unique values 0 missing | |

Z9 | numeric | 201 unique values 134 missing | |

Z10 | numeric | 154 unique values 0 missing | |

Z11 | numeric | 158 unique values 108 missing | |

Z12 | numeric | 295 unique values 106 missing | |

Z13 | numeric | 179 unique values 106 missing | |

Z14 | numeric | 146 unique values 136 missing | |

Z15 | numeric | 243 unique values 11 missing | |

Z16 | numeric | 48 unique values 2 missing | |

Z17 | nominal | 4 unique values 106 missing |

4

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

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

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

10.04

Third quartile of kurtosis among attributes of the numeric type.

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

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

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

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

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

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

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

2.72

Third quartile of skewness among attributes of the numeric type.

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

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

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

0.57

First quartile of kurtosis among attributes of the numeric type.

231.94

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

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

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

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

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

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

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

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

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

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

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

2.38

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

0.47

First quartile of skewness among attributes of the numeric type.

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

0.74

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

4.41

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

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

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

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

7.62

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

122.56

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

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

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

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

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

2.22

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

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

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

2

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

65.15

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

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