PSALSA Arguments - Teradata Vantage

Machine Learning Engine Analytic Function Reference

Product
Teradata Vantage
Release Number
8.00
1.0
Published
May 2019
Language
English (United States)
Last Update
2019-11-22
dita:mapPath
blj1506016597986.ditamap
dita:ditavalPath
blj1506016597986.ditaval
dita:id
B700-4003
lifecycle
previous
Product Category
Teradata Vantage™
SourceKey
Specify the source key (the names of the edges table columns that identify the target vertex). If you specify sources_table, the function uses only the vertices in sources_table as sources (which must be a subset of those that this argument specifies).

The function uses these names to construct the column names for recommended hub nodes in the output table.

TargetKey
Specify the target key (the names of the edges table columns that identify the target vertex). If you specify targets_table, the function uses only the vertices in targets_table as targets (which must be a subset of those that this argument specifies).
EdgeWeight
[Optional] Specify the name of the edges table column that contains edge weights. Each edge_weight is a positive value.
Default behavior: Each edge_weight is 1; that is, the graph is unweighted.
MaxHubNum
Specify the maximum number of hub vertices with the highest score output for each hub vertex. The max_hubs must be an INTEGER.
MaxAuthorityNum
Specify the maximum number of authority vertices with the highest score output for each hub vertex. The max_authority must be an INTEGER.
Accumulate
[Optional] Specify the name of the vertices table columns to copy to the output table.
TeleportProb
[Optional] Specify the probability ε, a DOUBLE PRECISION value between 0 and 1 that represents the jump to the seed vertex during random walk.
Default: 0.15
RandomWalkLength
[Optional] Specify the random walk length L for each hub vertex. L must be an INTEGER. The final hub/authority score is computed as ε*K v/L , where K v is the number of times that the random walk visits vertex v.
Default: 5000