KnowledgeSEEKER and KnowledgeSTUDIO users can use an optional In-Database Analytics feature in addition to performing analysis on Angoss data sets with data extracted from other sources.
In-Database Analytics performs data mining and predictive analytics directly on data stored in a database as opposed to working on a copy of the data. A key element of this process is that summary information only is extracted from the database, which is then used to drive many elements of the Angoss data mining functions.
In-Database Analytics offers many business performance improvements:
-
Duplication of data between the data warehouse and the analytical processing environment is eliminated.
-
Computation-intensive data mining algorithms (e.g., decision trees and data exploration) are performed on a well-tuned and managed database engine deployed on powerful servers.
-
Data security, integrity and standardization are maintained through one version of the data for all analytical and reporting functions.
-
There is no delay between data acquisition, preparation and analysis within the data warehouse.
Supported In-Database Analytics Features – Angoss’ In-Database Analytics driver supports KnowledgeSEEKER functionality and core KnowledgeSTUDIO features such as:
Data transformations and scoring and validation for Decision Trees and Strategy Trees are all enabled from within the database. Angoss enables complete analytics workflow within the database—from model development to deployment.
Angoss’ In-Database Analytics driver supports Teradata®, Microsoft® SQL Server and Netezza™ databases, with native open database connectivity (ODBC) drivers required for each of these databases.