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.

3.1.6 Gaussian Mixture Model (GMM)<br />

GMM is a Gaussian mixture model in which each component has its own weight, mean, and covariance matrix.<br />

Weight means the importance of a Gaussian distribution in the GMM, and mean and covariance matrix are the<br />

basic parameters of a Gaussian distribution, as shown in the following formula:<br />

Expectation maximization (EM) algorithm is used to inference all of the unknown parameters of GMM. The<br />

algorithm performs two steps: the expectation step and the maximization step.<br />

The expectation step calculates the contribution of training sample i to the Gaussian k:<br />

The maximization step calculates the parameters weight, mean, and covariance matrix:<br />

GMM can be used in image segmentation, clustering, and so on. It gives the probability of a sample belonging<br />

to each Gaussian component.<br />

Prerequisite<br />

●<br />

No missing or null data in the inputs<br />

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

<strong>PAL</strong> Functions P U B L I C 53

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

Saved successfully!

Ooh no, something went wrong!