Home
Name Modified Size InfoDownloads / Week
v2.8 2014-06-12
v2.1 2011-03-08
readme.txt 2014-06-12 1.2 kB
Totals: 3 Items   1.2 kB 0
--------------------------------------
Release Notes 12/06/2014, build 2.8.28
--------------------------------------
- Added support for ordinal (categorical attributes where order is significant) and date attributes. These two new attribute types can be scaled in the same manner that continuous attributes are scaled in FcaBedrock
- Added 3 binning methods which can be used to automatically create ranges for continuous attributes during autodetection:
	* Equal width - the attribute is scaled using ranges of equal width, based on the min/max values of the attribute
	* Equal frequency - the attribute is scaled in such a manner so that each range contains (approximately) the same number of values
	* Standard deviation - attribute is scaled up to 3 standard deviations
- Added 'First row as headers' option - use this when first row of your flat-file csv contains column names as it's first row
- Added 'Inconsistency Mode' - Objects are 'inconsistent' when they have more than one value for a particular attribute. During this mode inconsistent objects, along with the attributes which render the object inconsistent, are converted into a formal context. 
- Minor updates to help file
- Stability bug fixes
Source: readme.txt, updated 2014-06-12