6. Machine Learning for Servicing Queue Creation

OFSLL is equipped to leverage Oracle Data Mining capability to give additional Machine Learning features. Oracle Data Mining provides a powerful, state-of-the-art data mining capability within Oracle Database. Machine Learning capability is leveraged to identify the Queue/Segmentation that can be created for the Account data. Intelligent Segmentation feature provides list of Clusters/Queues using the Machine Learning Algorithm for a given account condition.

Orthogonal Partitioning Clustering (O-Cluster), an Oracle-proprietary Clustering algorithm, has been used to create Intelligent Segments/Clusters for a given condition. Clustering Algorithm will discover natural groupings in given data.

This helps to automate the manual process of queue creation which is otherwise done by identifying different segments of Accounts and assigning day to day Customer Service Activities.

The Intelligent Segmentation screen in OFSLL UI is developed using the Oracle JavaScript Extension Toolkit (Oracle JET) frame work. For information on usability, refer to ‘Intelligent Segmentation’ section in setup guide.

To enable Machine Learning Service Queue creation in OFSLL:

1. Provide Grant Mining Model privilege to the schema user.

GRANT CREATE MINING MODEL TO <Schema>;

2. Execute the sql script ‘upgrade_ML_dataset.sql’ (Available in data fix folder of release bundle) to load service account's data on to ML table.

Following is a known issue:

• External tables are not applicable in ATP-D

Note

Ensure that customer service accounts exist in the system so that based on the data pres­ent in accounts, account conditions and assets tables gets loaded on to ML table.

3. Enable menu access for ‘Intelligence Segmentation’ screen with key column data ‘FLL.SET.JET.INTELLIGENTSEGMENTATION.MENU|SETUP_USER_MENU’ from seed data setup screen.

For more information on Oracle Data Mining, refer to

https://www.oracle.com/database/technologies/advanced-analytics/odm.html