Understanding Data Science Setup

The Data Science Setup for PeopleSoft applications automates the process of creating and running data science jobs, so that you can schedule the jobs to run periodically. PeopleTools also automates the fetching of data science prediction endpoints, enabling you to adapt to the new features delivered by OCI data science.

The process flow involves the following steps:

  1. PeopleSoft sends data to Elasticsearch using Direct Transfer.

  2. OCI Data Science service pulls data from Elasticsearch.

  3. The data is used to train the machine learning model using the Python file artifact on the OCI Data Science service.

  4. When the model is created, it can be saved in the OCI object storage.

  5. The saved model can be deployed using PeopleSoft Data Science Setup.

  6. OCI Data Science service pulls credentials and job information from PeopleSoft to automatically run jobs.

  7. OCI Data Science runs prediction endpoint using the model name and data set.

The following diagram illustrates the process flow among PeopleSoft, Elasticsearch, and OCI Data Science service.

The diagram illustrates the process flow among PeopleSoft, Elasticsearch, and OCI Data Science service.

 Implementation Process Flow

PeopleSoft provides a navigation collection – Data Science Setup – that enables you to perform the following tasks:

  • Set up data source.

  • Specify OCI Data Science server.

  • Build an index in Elasticsearch.

  • Register a model.

  • Get OCI credentials and job information.

  • Schedule the Application engine program.