2.14.2 Description

This topic provides a description of the machine learning for servicing queue creation.

The New Setup Menu is introduced in the User “Intelligent Segmentation”. User will have the option to select the Company, Branch, and Account Condition.

Cluster:

When the user Submit, the System will create the Clusters and display them on the user interface below. The screen will display the record count for each Cluster ID. The size of Clusters on the screen will change according to the size of the record counts.
Description of cluster.png follows
Description of the illustration cluster.png

Create:

Clicking on the Create will create the Queue in Oracle Financial Services Lending and Leasing in a Disabled status, and the selection criteria of the Queue will be populated with the Cluster Attributes.

Click on a Cluster, and the system displays the criteria for cluster creation. The Oracle clustering algorithm uses attributes to create the cluster; these attributes can be viewed by clicking on the clusters. This will give more information about attribute names and attribute values for the given Cluster ID. Each Cluster ID will have its own set of rules.
Description of create.png follows
Description of the illustration create.png

Segment:

Will display another visual representation of the Clusters. The screen will display the record count for each Cluster ID.


Description of segment.png follows
Description of the illustration segment.png

Select a cluster and system that gives the capability to create a Cluster as a Queue with the Criteria.

Note:

  • Machine Learning attributes are directly inserted as queue criteria; therefore, the user is expected to update the selection criteria in the queue and ensure the formula is formatted correctly to enable the queue.
  • To utilize the Oracle Machine Learning capability, a financial institution is recommended to run this on a non-production / data warehouse database.

Insights:

Another visual representation of the Clusters identified by the Machine Learning. The clustering algorithms supported by Oracle Data Mining perform hierarchical clustering. Before arriving at the final cluster, it divides the data into various intermediary clusters. The leaf clusters are the final clusters generated by the algorithm.
Description of insight.png follows
Description of the illustration insight.png

Select a cluster and system that gives the capability to create a Cluster as a Queue with the Criteria.

Machine Learning Data Set Creation Through Batch Job:

Existing batch job leveraged to load machine learning data set, which is input for cluster creation.
  • SET-QCS à QMLPRC_BJ_100_03 à ML DATASET CREATION FOR INTELLIGENT SEGMENTATION.
  • System refers to the columns from User Defined Tables.