1.3 Component Details

The following table lists the component details.

Table 1-1 Component Details

Component/Service Details
Compliance Studio UI You can access the Compliance Studio UI via browser and enter the login credentials along with the language. For valid login credentials, it navigates to the Workspace Summary page.
Notebook Server UI You can access Notebook Server UI through the Compliance Studio UI.
Spark Interpreter You can connect to a big data cluster and create the models to perform analytics on data present in the Big data clusters.
Python Interpreter You can create/execute the Python models using this Interpreter. Analytics can be done with any python library.

By default, python interpreters are configured with predefined Conda environments as follows:

  • default_8.1.2.9.0
  • ml4aml_8.1.2.9.0
  • sane_8.1.2.9.0

For more information, see the OFS Compliance Studio Administration andConfiguration Guide.

JDBC Interpreter You can create/execute the SQL models using this Interpreter. By default, this is connecting to Studio schema.

You can connect to any schema by changing the interpreter configuration. For example, BD or ECM schema.

NOTE:

This feature is not recommended approach because it can only be used to connect to a single schema, and all users will have access to that, rather than access being managed per user. In future releases this interpreter will not be enabled by default but instructions will be given to enable if required.

Limitation

  • Data source configuration is not dynamic; instead, it is static from the Interpreter Configuration screen.
  • There is no restriction or secure access of data provided with this interpreter.

Recommendation

Users are recommended to use a python interpreter to get dynamic data source configuration; even data access permission features can also be used with this interpreter.
PGX Interpreter
  • pgx-java: Java-based Interpreter, you can create/execute Java- based models and interact with the PGX server for graph analytics.
  • pgql:SQL is like an interpreter to query on the graph.
  • pgx-python:python based Interpreter with a PGX python client embedded in it to query on graph present in the PGX server.
  • pgx-algorithm:Graph toolkit that provides a graph query language and optimized analytics algorithms.
Meta Service This service is responsible for setting up metadata related to Compliance Studio in Studio Schema.
Load to OpenSearch This service manages OpenSearch indexes used to resolve the entity based on the matching rules.
Batch Service This service is responsible for executing some of the batch jobs of Compliance Studio.

For example, ETL for graph analytics or entity resolution.
Entity Resolution It is responsible for resolving entities using matching and merging rules.
  • Graph ER: It creates Similarity Edges between nodes by comparing the attributes of the nodes and identifying where the similarity is significant enough to create an edge so the nodes are linked with the graph model and can be analyzed as a single entity.
  • Global Party ER: It creates a Global Party of similar entities by comparing multiple attributes of entities using matching and merging rules.

    For more information on merging and matching rules, see OFS ComplianceStudio Matching Guide.

Matching Service It is responsible for scoring in ER based on matching rules. It has the following scoring methods:
  • Jaro-Winkler
  • ML-BoostedName
  • ML-BoostedAddress
  • Levenshtein
  • Individual
  • Name
  • Entity
  • Name
  • Default

For more information on merging and matching rules, see the OFS Compliance Studio Matching Guide.

Graph Service This service is used for managing graphs in Compliance Studio.
Pipeline Service This service is used for extract transform and load data into target tables.
ML Model Templates The prepackaged Model templates allow you to perform the following:
  • Model segmentation (model grouping)
  • Load and Preview data
  • User-defined transformation (deriving additional features)
  • EDA
  • Feature selections
  • Model training
  • Evaluation
  • Model Agnostics (Explainability)
  • On-going validations
Python Python contains all packages required to execute ML4AML models. For example, scikit-learn pandas.
MMG Service This service is used to manage the following functions:
  • Work spaces and sandbox
  • Data sources (external, local file, relational, and distributed)
  • Model complete life cycle, governance, and execution
  • Batch and Scheduler services
  • User roles and accesses
  • User Provisioning and authentication
Parallel Graph Analytics Server Graph analysis lets you reveal latent information that is not directly apparent from fields in your data but is encoded as direct and indirect relationships - metadata - between elements of your data. This connectivity-related information is not apparent to the naked eye but can have tremendous value when uncovered. PGX is a toolkit.

For graph analysis, It supports both efficient graph algorithms and fast SQL- like graph pattern matching queries.

FCGM is loaded into the PGX server for analytics.
BDPACK In Compliance Studio, the graph model is based on the BD Pack’s FCDM model and ML4AML using the same data model.

For more information, see the Behavior Detection Application Pack.

ECMPACK In Compliance Studio, the graph model is based on the ECM Pack’s FCDM model.ECM is also used to correlate events generated via Compliance Studio and for case investigation.

For more information, see the EnterpriseCase Management ApplicationPack.

Oracle DB Compliance Studio stores the metadata in the Oracle DB.
Investigation Toolkit OFS Investigation Toolkit is an application built on Compliance Studio, allowing investigators to view the case and adhoc information within, then creates case narratives and insights, highlight risk factors and red flags meaningful to the investigation, and recommend actions based on graph scoring algorithms.

For more information, see the Investigation Toolkit.

Scenario Conversion Utility This utility converts the Behavior Detection scenario into Compliance Studio scenario.
Identity Provider Identity Provider (IDP) is required for SSO/SAML authentication.
OpenSearch Cluster An OpenSearch cluster is a group of nodes that have the same cluster name attribute. As nodes join or leave a cluster, they reorganize to evenly distribute the data across the available nodes. If you are running a single instance of OpenSearch, you have a cluster of one node.

It is used for a matching service engine used for Entity Resolution and Similarity Edge for Graph Nodes.