|
|
Financial Services Crime and Compliance Studio Release 8.0.8.2.0 |
Oracle Financial Services Crime and Compliance Studio is a separately licensed product.
Description
ID 32342898 (Kubernetes) and 32358163 (On premise) – OFS Crime and Compliance Studio Release 8.0.8.2.0
This release of OFS Crime and Compliance Studio includes the enhancements and bug fixes since the 8.0.8.1.0 release.
New
Features in this Release
The following new features are incorporated in this release. For detailed information on how to use these features, see the Oracle Financial Services Crime and Compliance Studio Documentation.
1. The FCC Studio can be integrated with Model Management and Governance Application (MMG).
· The FCC Studio Integration with MMG provides a seamless integration and eliminates the requirement to login separately to FCC Studio. You can also work with your Notebooks from MMG or vice versa. For more information, about the MMG Integration parameters, see Oracle Financial Services Crime and Compliance Studio Installation Guide 8.0.8.2.0.
2. Infer and Training API
· Infer and Training APIs can be used to train the machine learning model with your data based on the existing data. For this release, this is limited for name matches.
3. New Scoring Methods when creating your Rulesets.
You can use the following Rulesets when creating a ruleset:
· ML-Boosted Name Matching: Used maching learing model to identify similarity between 2 strings.
· Reverse Jaro: Identify similarity between strings using a tokenized approach.
· Fuzzy Tokenized: Identify similarity between tokens of a strings in a matrix based approach.
4. Updates Manual Decisioning feature in FCC Studio.
· When the approving or rejecting the Similarity Edges, you can now provide comment for the selected similarity edge, using the Comments drop-down. Then you can select Do Names Match? drop-down and select Yes or No.
· Based on the selection, the similarity edge is stored in the fcc_er_ml_training_data_name_boosted table, with the source MANUAL DECISION. If Yes is selected the MATCH value is 1 and if No is selected then MATCH value is 0.
How
to Install this Release
For detailed instructions on installing this OFS Crime and Compliance Studio Release, see the OFS Crime and Compliance Studio Application Pack Installation Guide 8.0.8.2.0.
Known Issues or Limitations in This Release
The following table lists the known
issues or limitations in this release.
|
Sl. No. |
Bug ID |
Known Issue or Limitation |
Workaround |
|
1 |
32360369 |
If a shared notebook from your admin is accessed. You will see the following: ERROR - NOTAUTHORIZEDORNOTFOUND
POPS EACH TIME USER OPENS A SHARED NOTEBOOK |
Close the pop-up to see the contents of the
notebook. |
|
2 |
32360382 |
No appropriate messages, in case of invalid data in the console, when user executes FCCM_Studio_ML_Model_Training.sh file. For example: column “MATCH” has value 0 or 1 for all the records. |
|
|
3 |
32360387 |
ML Scoring Method will give high scores for non-matching or partial matching names |
|
|
4 |
32360399 |
When user starts on premise matching service, then the logs related to kubernetes is populated. |
You can ignore the logs. |
|
5 |
32360405 |
On prem - Exception messages in the console, when starting the FCC STUDIO. |
You can ignore the message. |
|
6 |
32355011 |
I FRAME ISSUE - From MMG, user can open notebooks and navigate to the other pages in FCC Studio |
|
|
7 |
32339325 |
WORKSPACE dropdown is visible for Notebooks that are not applicable for MMG. For Example: Scenario Notebooks |
|
Oracle
Financial Services Software | Copyright © 2020, Oracle
and/or its affiliates. All rights reserved. | Phone:
+1.650.506.7000 | Fax: +1.650.506.7200
| www.oracle.com/industries/financial-services/index.html