The FrequentPaths function requires an input table, which contains the sequence data to process. The input can be in either of these formats:
- Sequence/path format:
Each row contains a string in the format '[item[, ...]]', where the outer brackets belong to the string (for example, '[A, B, C, D]'). To output strings in this format, you can use the function nPath with its Accumulate argument.
- Item format:
Each row represents one item in a sequence. With this format, you must specify either the ItemColumn or ItemDefinition argument.
If the input table does not have an item column (specified by the ItemColumn argument), the function also requires an item definition table (specified by the ItemDefinition argument).
Column Name | Data Type | Description |
---|---|---|
partition_column | INTEGER, SHORT, or LONG | Sequence index. Rows with the same index belong to the same sequence. |
item_column | CHAR, VARCHAR, or TEXT | Sequence item. The input table has this column only if you do not specify an item definition table. |
time_column | Any except DOUBLE PRECISION | Optional. Time stamp of sequence item. Items in the same sequence with the same time stamp belong to the same item set. |
path_column | CHAR, VARCHAR, or TEXT | Optional. Paths in the form of sequence strings. |
Column Name | Data Type | Description |
---|---|---|
index_column | INTEGER, SHORT, or LONG | Used to determine when more than one predicate in definition_column is satisified, in which case item_column applies. |
definition_column | CHAR, VARCHAR, or TEXT | Predicate definition. |
item_column | CHAR, VARCHAR, or TEXT | Sequence item for which the predicate is true. The function applies the predicates to the input table in index order. If more than one predicate is true for a row, the function assigns the row the value that corresponds to the predicate with the smallest index. If an input table row has no corresponding definition in the item definition table, the function skips that row. |