Operators in StreamBase are components that perform a specific type of runtime action, such as filtering, merging, or querying, on the data streaming through them. You can separately configure each instance of an operator in StreamBase Studio using its Properties view.
The Dynamic Learning operators enable statistical and predictive analytics computed from streaming data, with results tables and statistics themselves delivered downstream as data streams. This will enable downstream consumers such as Spotfire to visualize in real-time results, such as correlations, feature importance (in some predictive analytics problem), or just standard Six-Sigma-type statistical summaries as they are commonly computed in manufacturing.
All the Dynamic Learning Operators have a dependency on the SMILE library to work.
Contents
- Using the ANOVA Operator
- Using the Chi-Square Test Operator
- Using the Classification Trees Operator
- Using the Correlations Operator
- Using the Descriptive Statistics Operator
- Using the Frequency Tables Operator
- Using the Kolmogorov-Smirnov Test Operator
- Using the Linear Regression Operator
- Using the Logistic Regression Operator
- Using the Multilayer Perceptron Classification Operator
- Using the Multilayer Perceptron Regression Operator
- Using the Paired T-Test Operator
- Using the Regression Trees Operator
- Using the Single Sample T-Test Operator
- Using the SVM Classification Operator
- Using the SVM Regression Operator
- Using the Two Sample T-Test Operator
- Using the T-Test By Groups Operator