25.10.2016 Views

SAP HANA Predictive Analysis Library (PAL)

sap_hana_predictive_analysis_library_pal_en

sap_hana_predictive_analysis_library_pal_en

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

DROP TABLE <strong>PAL</strong>_C45_TREEMODEL_TBL;<br />

CREATE COLUMN TABLE <strong>PAL</strong>_C45_TREEMODEL_TBL LIKE <strong>PAL</strong>_C45_TREEMODEL_T;<br />

DROP TABLE <strong>PAL</strong>_C45_PMMLMODEL_TBL;<br />

CREATE COLUMN TABLE <strong>PAL</strong>_C45_PMMLMODEL_TBL LIKE <strong>PAL</strong>_C45_PMMLMODEL_T;<br />

CALL "DM_<strong>PAL</strong>".<strong>PAL</strong>_C45_PROC(<strong>PAL</strong>_C45_DATA_TBL, #<strong>PAL</strong>_CONTROL_TBL,<br />

<strong>PAL</strong>_C45_TREEMODEL_TBL, <strong>PAL</strong>_C45_PMMLMODEL_TBL) with OVERVIEW;<br />

SELECT * FROM <strong>PAL</strong>_C45_TREEMODEL_TBL;<br />

SELECT * FROM <strong>PAL</strong>_C45_PMMLMODEL_TBL;<br />

Expected Result<br />

<strong>PAL</strong>_C45_TREEMODEL_TBL:<br />

<strong>PAL</strong>_C45_PMMLMODEL_TBL:<br />

3.2.4 CART Decision Tree<br />

Classification and regression tree (CART) is used for classification or regression and only supports binary split.<br />

CART is a recursive partitioning method similar to C4.5 decision tree. It uses GINI index or TWOING for<br />

classification, and least square error for regression. In <strong>PAL</strong>, CART only supports the GINI split strategy. The<br />

surrogate split method is used to support missing values when creating the tree model.<br />

Prerequisites<br />

●<br />

●<br />

The target column of training data must not have null values, and other columns should have at least one<br />

valid value (not null).<br />

The table used to store the tree model is a column table.<br />

144 P U B L I C<br />

<strong>SAP</strong> <strong>HANA</strong> <strong>Predictive</strong> <strong>Analysis</strong> <strong>Library</strong> (<strong>PAL</strong>)<br />

<strong>PAL</strong> Functions

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!