<< Click to Display Table of Contents >> Current:  Advanced Analytics > Statistical Analysis  Analysis of Correlation

Analysis of Correlation (correlation analysis) is to study whether there is a certain dependence relationship between phenomena, and to explore the relevant direction and degree of correlation of the specific dependence relationship. It is a statistical method to study the correlation between random variables. There are three calculation methods for correlation measure: Pearson, Kendall, Spearman.

 

How-To-Use:

Please select two or more numerical features as input data. After setting the correlation analysis, you can view the correlation matrix by connecting the table view, and view the matrix diagram by connecting the picture view.

 

Precautions:

You cannot select the string field.

 

❖Configuration

After adding the Correlation analysis node to the experiment, you can set the Correlation analysis node through the "Configuration" page on the right.

[Method] Measure the strength of the linear correlation between two variables; it is achieved on the basis of variance and covariance, and is sensitive to outliers; it obeys normal distribution or unimodal distribution close to normal distribution; Pearson data demand: two Variables are continuous value data. Spearman: Measures the strength of the connection (change trend) between change variables; it is obtained on the basis of the relative size of the rank (ranking), which is more robust to outliers; Spearman Data Demand: Both variables are continuous data or graded data. Kendall: Obtained based on the idea of collaborative filtering, to measure the trend of synergy between variables; robust to outliers; Kendall Data Demand: both variables are continuous data or grade data

[Operation column] Select a numeric field, no less than two fields.

 

Right-click menu of analysis of correlation:

 

❖Run Analysis of Correlation Node

Run the node, pass the data to DM-Engine for calculation, and get the output result.

 

❖Reset Analysis of Correlation Node

The node that has been running is reset, the returned result is deleted, and the node status is changed to not running.

 

❖Rename Analysis of Correlation Node

In the right-click menu of the Analysis of Correlation node, select "Rename" to rename the node.

 

❖Refresh Analysis of Correlation Node

In the right-click menu of the Analysis of Correlation node, select "Refresh" to update the synchronization data or parameter information.

 

❖Save as Composite Node

In the right-click menu of the Analysis of Correlation node, select "Save as Composite Node",The selected node can be saved as a composite node to realize a multiplexing node, and the parameters of the saved node are consistent with the original node.

 

❖Cut Analysis of Correlation Node

In the right-click menu of the Analysis of Correlation node, select "Cut" to realize node cutting operation.

 

❖Copy Analysis of Correlation Node

In the right-click menu of the Analysis of Correlation node, select "Copy" to realize node  replication operation.

 

❖Delete Analysis of Correlation Node

In the right-click menu of a Analysis of Correlation node, select "Delete" or click the delete key on the keyboard to delete the node and its input and output connections.