Shuttle
Control
|
§ Back § |
Attributes:
The shuttle dataset contains 9 attributes all of
which are numerical.
Instances: 43500 items
Classes:
Approximately 80% of the data belongs to class 1.
Therefore the default accuracy is about 80%.
The aim here is to obtain an accuracy of 99 - 99.9%.
Of its seven clusters, one is very large with
approximately 80% data-items, two are moderately
large with approximately 15% and 4% data-items,
respectively, one is fairly small with aproximately
0.8% , and the remaining three are extremely small
with just 0.09%, 0.01% and 0.03% data-items,
respectively.
Vista visualization: 1745 unseparated
items(4%)
* Can't separate some small clusters well.