22.0.0

**Table 1** Features and documents related to 22.0.0
Feature	Document	Description
Machine learning algorithms	Machine Learning Algorithm Library Release Notes	Updated the machine learning algorithm library to 2.2.0. Adapted the Decision Tree, Random Forest, and GBDT algorithms to Spark 3.1.1. Added the DTB and Word2Vec algorithms (adapted to Spark 2.3.2, Spark 2.4.5/2.4.6, and Spark 3.1.1).
	Machine Learning Algorithm Library Version Mapping
	Machine Learning Algorithm Library Feature Guide
	Machine Learning Algorithm Library Tuning Guide
	Machine Learning Algorithm Library Acceptance Test Guide
	Machine Learning Algorithm Library Developer Guide
OmniData	OmniData Communication Matrix	Upgraded OmniData to 1.4.0 and changed the deployment mode to client/server decoupling. Optimized Hive engine performance in scenarios where the Filter operator is used. Upgraded HAF to 1.3.0 and supports independent deployment.
	OmniData Release Notes
	OmniData Version Mapping
	OmniData User Guide
OmniOperator	OmniOperator Release Notes	Added UDF support. Implemented vectorized instruction acceleration. Added support for overflow processing of the decimal type and conversion of the cast type in expressions. Enhanced basic Spark functions, aggregation operators, and Join operators. Adapted to Spark AQE. Added support for the Short type in OmniVec.
	OmniOperator Version Mapping
	OmniOperator User Guide
OmniCache Materialized View	OmniCache Release Notes	The OmniCache materialized view feature uses AI algorithms to recommend the optimal materialized view from historical SQL queries, automatically matches user SQL statements with a materialized view in Spark, and applies the matched materialized view to the execution plan of SQL statements. This feature reduces repeated calculations and increases the query efficiency. This issue is the first official release.
	OmniCache Version Mapping
	OmniCache User Guide