User's Manual

11
New Features
The Netezza Time Series node analyzes time series da t a and can predict future
behavior from past events.
The Netezza Generalized Li near model expands the linear regression model so that
the dependent variable is related to the predictor variables by m eans of a s pecified
link function. Moreover, the m odel allo w s f or th e dependent variable to have a
non-normal distribution.
IBM SPSS Modeler Premium
The EA Export node is a terminal node that reads entity data fro m a data source and
exports the data to a repository for the purpose of entity resolution.
The Entity Analytics(EA) sou rce no de reads the resolved entities from the repository
and passes t h
is data to the stream for further processing, such as formattin g into
a report.
The St reaming EA node compares new cases against the entity data in the rep osit ory.
The SNA Group Analysis no de builds a model of a social network based on i nput
data about the social groupings within the network. This technique identifies links
between the group memb ers , and analyzes the interactions within the groups to
produce key performance indicators (KPIs). The KPIs can be used for purposes such
as churn prediction, anomaly detection, or group leader identification.
The SNA D i ffusion Analysis node models the flow of information from a group
member to their social environment . A group member is assigned an initial weighting,
which is propagated across the network as a gra dually reducing figure. This process
continues until each member of the network has been assigned a weighting relative to
the original group member, according to the amount of information that has reach ed
them. The individual member scores are then derived directly from these weightings.
In this way, for example, a service provider could identify customers that are at a
higher risk of churn according to their relationship with a recent churner.