Data Mining update in Oracle

Abstract: Oracle Data Mining (ODM) is a data mining and prediction analysis engine in a database, allows you to create and use advanced predictive analytics models on data that can be accessed through your Oracle Data Infrastructure.

I recently got an Oracle Data Mining (ODM) update from Oracle. Oracle Data Mining (ODM) is a data mining and prediction analysis engine in a database, allows you to create and use advanced predictive analytics models on data that can be accessed through your Oracle Data Infrastructure. Last year, the Oracle Data Mining (ODM)ArticleSince then, they have released Oracle Data Mining (ODM) 11.2.

Of course, the basic structure of Oracle Data Mining (ODM) has not changed. It is still a "out-of-Database" solution that appears to be executed in the database through SQL and PL-SQL APIs. It has 12 typesAlgorithmAnd more than 50 statistical functions, and complete modeling and scoring in the data. Oracle Text Functions are integrated so that text mining algorithms can use these functions. In addition, because Oracle Data Mining (ODM) is used to mine data in star mode, it can process unlimited input of attributes, transaction data and unstructured data, such as clobs, tables, or views.

This version has the preview graphical user interface (GUI) discussed last time and is officially released. This new graphical user interface (GUI) is an extension of SQL developer 3.0 (free to use and downloaded by tens of thousands of SQL/database personnel. The "classic" interface (wizard-based access to APIs) is still available, but the new interface is more in line with the current technical level of the analysis tool.

The graphical user interface (GUI) allows you to manage connections to databases in projects with sequential workflow control. These look very similar to the SAS enterprise mining or ibm spss modelers with various action nodes that can be connected to the data mining stream. Workflows can access tables, connections, views, and Remote Data managed by Oracle databases in real time. Other nodes include those that create views, browse data, merge and cleanse data, convert and aggregate data, and create models. A filter node allows you to remove error data elements, unimportant attributes, sample data, and more. Statistics required for all analysis and report nodes are also run in the database and all steps in the workflow generate an SQL auxiliary graph.

Nodes such as classification can run in multiple models at the same time, and by default, they are not allowed to run in different model methods. Some advanced user functions allow slight parameter adjustment. You can select one or more models to advance through the workflow. Once a model is ready, it can be applied to a database table or embedded in the database as an instant column. The exadata model is used to promote the storage level to facilitate the execution of the best results.

Extended features include training "Tips" and sample data that help data reporters use tools to attach, and an embedded "test product advertisement" to database administrators (DBAs) other developers who are using SQL introduce the concept of data mining. More and more training courses are available online. One of the most meaningful factors in Oracle Data Mining (ODM) is that it can be used in other Oracle solutions. It is embedded as a pre-node in a variety of application systems-from fusion Human Capital Management to Oracle Business Intelligence Enterprise Edition (obiee ). There are also some ongoing work on Oracle Real-time decision-making (RTD. This factor, coupled with the shared business intelligence (BI) sales team, is increasingly occupying and broadening the scope of Oracle Data Mining (ODM), both in Oracle and its customer base.

In-database analysis is currently a hot topic, and Oracle Data Mining (ODM) is capable of fully developing and executing models in the database, make it the best choice for anyone interested in using predictive analysis to take advantage of most of their operational data.