Oracle Machine Learning

In-database algorithms

Data scientists can easily build and deploy high-performance machine learning models with more than 30 scalable algorithms built into Oracle Database.
Explore machine learning techniques and support algorithms (PDF)

No data movement

Data scientists, and developers are able to process data where it resides to help simplify model building and deployment, reducing application development time, and helping ensure data security.

Single source of truth

Data scientists can maximize the timeliness and relevance of machine learning processes by analyzing current business data inside the database.

Automatic data preparation

Data scientists are able to accelerate time to insight by automating commonly required transformations for in-database machine learning algorithms.

Rapid enterprise deployment

Developers achieve immediate machine learning model availability with simple SQL and easy deployment options using representational state transfer (REST) interfaces.

Oracle Machine Learning Notebooks

Data scientists, and developers use an easy-to-use, interactive multiuser collaborative interface based on Apache Zeppelin notebook technology, supporting availability for SQL, and PL/SQL interpreters for Oracle Autonomous Database.

• Oracle Machine Learning Notebooks with Autonomous Data Warehouse documentation
• Oracle Machine Learning Notebooks with Autonomous Transaction Processing documentation
• Oracle Machine Learning Notebooks with Autonomous JSON Database documentation

Oracle Machine Learning for SQL

Data scientists are able to simplify the creation of machine learning models using familiar SQL and PL/SQL for data preparation, and machine learning model building, evaluation, and deployment inside Oracle Database.

• Oracle Machine Learning for SQL documentation
• Technical brief (PDF)

Oracle Machine Learning for R

R users gain the performance and scalability of Oracle Database for data exploration, preparation, and machine learning from a well-integrated R interface which helps in easy deployment of user-defined R functions with SQL on Oracle Database.

• Install Oracle Machine Learning for R
• Technical brief (PDF)

Oracle Data Miner

Data scientists and data analysts can use SQL Developer add-in drag-and-drop interface to explore data and build analytical methodologies (workflows) that can be shared and scheduled to solve data-driven problems.

• Install Oracle Data Miner (PDF)
• Oracle Data Miner documentation

Oracle Machine Learning for Spark

Data scientists are able to use all nodes of a big data cluster with scalable Spark-based algorithms on data from Hive, Impala, HDFS via an R API for faster model building and data scoring.

• Install Oracle Machine Learning for Spark
• Apache Hive and Impala support (PDF)

High performance compute

Data scientists are able to avoid performance issues during data preparation, model building, and data scoring using the built-in parallelism and scalability of Oracle Database, with optimizations for Oracle Exadata and other environments.

Converged database

Developers can easily deploy Oracle Machine Learning in diverse applications using multiple types of data, such as spatial and graph data, by leveraging converged Oracle Database capabilities.

In-database scoring

Data scientists, and developers can rapidly score large volumes of data using Exadata “smart-scan” technology to deliver faster results.

Security

Data scientists and developers using Oracle Machine Learning are protected with Oracle Database built-in security, encryption, and role-based access to user data and Oracle Machine Learning in-database models.