Data Transformation

<< Click to Display Table of Contents >>

Current:  Appendix > Page Introduction > Advanced Analytics Page > Advanced Analytic Editing Page > Edit Area > Node Type 

Data Transformation

Previous pageReturn to chapter overviewNext page

Data Transformation

The data transformation refers to the plug-in operator of Y-AI, including two operators of Normalization and Join.

Normalization

Normalization is mainly used to eliminate the influence of unit and scale differences between features.

Parameter list: Change the normalization method used for column standardization by selecting parameters.

Output list: data set

Drag and drop a data set view to connect the sampling nodes and select the columns to be standardized. After running the experiment, you can get the normalized columns and other columns of the pre-data node in the data set view.

ML160

After normalization the experiment, you can get the data listed after Normalization.

ML161

Join

Combine the output data sets of multiple nodes into one data set; similar to the join operation of multiple tables in a database. For specific usage, refer to the Advanced analysis module.

Parameter list: By changing the parameters, the combined output result column and processing method can be changed.

Output list: data set

Note: The front node can only be connected to the plug-in operator node. It is often used before the voting node to combine the prediction data of multiple models and connect the voting node to elect the best result.

ML162