All output files are text-based

Config Files
.config files contain a textual representation of the parameters used for the transactions, taxonomy or sequences.

Pattern Files
.pattern files contain two sections, items and patterns.

Each item is listed on a separate line and contains the following fields separated by a space

Id Item id
Probability Item probability
First Child Item id of first child (taxonomies only)
Last Child Item id of last child (taxonomies only)

Each pattern is listed on a separate line and contains the following fields separated by a space

Probability Pattern probability
Confidence Pattern confidence
Items... Item ids...


Transaction Files
Each item generated as part of a transaction is listed on a separate line and contains the following fields separated by a space

Transaction Id The id of the transaction to which the item belongs
Item Id The id of the item


Taxonomy Files
Taxonomy files specify the structure of a generated taxonomy. Each line specifies two fields

Item Id The id of the item
Parent Id The id of the item's parent


Sequence Files
The format of sequence files depends on whether or not the flat format option is used

Flat sequence files list each sequence on a separate line, with items separated by commas and elements by semi-colons.

Standard sequence files list each generated item on a separate line and contains the following fields separated by a space

Sequence Id The id of the sequence to which the item belongs
Transaction Id The id of the transaction to which the item belongs
Item Id The id of the item

Last edited Feb 9, 2011 at 9:54 AM by arthur_pitman, version 6

Comments

No comments yet.