Machine Learning for Predictive Analytics
Vertica provides a number of machine learning functions for performing in-database analysis. These functions can perform a number of data preparation and predictive tasks—for example:
- Balance an uneven distribution of classes — See Balancing Imbalanced Data for an example.
- Remove outliers from your data — See Detecting Outliers for an example.
- Impute missing values in a data set — See Imputing Missing Values for an example.
- Normalize data to organize different scales of numeric data to an equivalent scale — See Normalizing Data for an example.
- Create a sample of a larger data set — See Sampling Data for an example.
- Use regression algorithms to make predictions about features in your data set and an observed value response — See Regression Algorithms for more information.
- Use classification algorithms to assign items in a data set to different categories — See Classification Algorithms for more information.
- Use the clustering algorithm to partition data using k-means clustering — See Clustering Algorithms for more information.
For more information about specific machine learning functions see Machine Learning Functions.