Career Planning position recommendation setup

The position recommendation solution is part of the Career Planning functionality in HR Talent. To use this feature, enable Career Planning and define position and job qualification requirements.

Employees must also declare their qualifications in the application. This information is used to compare employee qualifications with position and job requirements.

Before you can use Career Planning position recommendations in the Employee space, administrators must complete configuration tasks. These tasks include data replication, machine learning training, and workflow setup.

To set up the position recommendation capability, you must have these security roles:

  • HRAdministrator
  • Employee
  • IONAPI-Administrator
  • IONDeskAdmin
  • DATAFABRIC-User

You must also have access to Infor AI Platform functionality, such as dataset, quest, and endpoint management.

Data replication to Data Lake

The Career Planning position recommendation feature uses data that is replicated to Data Lake. The required business classes are included in the GHR_ST replication set.

Use Landmark Replication Sets to publish required business class data. The delivered GHR_ST replication set includes these business classes:

  • Position
  • PositionCompetencyAndSkill
  • PositionCredential
  • PositionEducation
  • PositionExperience
  • Job
  • JobCompetencyAndSkill
  • JobCredential
  • JobEducation
  • JobExperience

These business classes require these fields:

  • Position: HROrganization, Position, HROrganizationUnit, PositionCategory, PositionSubCategory, PositionIsSupervisor, Location
  • PositionCompetencyAndSkill: HROrganization, CompetencyAndSkill, EssentialRatingLevel, Active
  • PositionCredential: HROrganization, Credential, Essential, Active
  • PositionEducation: HROrganization, Education, Essential, Active
  • PositionExperience: HROrganization, Experience, YearsExperience, Active
  • Job: HROrganization, Job
  • JobCompetencyAndSkill: HROrganization, CompetencyAndSkill, EssentialRatingLevel, Active
  • JobCredential: HROrganization, Credential, Essential, Active
  • JobEducation: HROrganization, Education, Essential, Active
  • JobExperience: HROrganization, Experience, YearsExperience, Active

See Infor Landmark Technology Library (Cloud) and select Administration > System Administration > Data replication > Creating and running replication sets > Starting and scheduling data replication.

As an administrator, ensure that required tables are available in Data Lake by using the delivered replication tables for position and job qualification data.

You can use these replication names in the replication set to verify data:

  • HCM_Position
  • HCM_PositionCompetencyAndSkill
  • HCM_PositionCredential
  • HCM_PositionEducation
  • HCM_PositionExperience
  • HCM_Job
  • HCM_JobCompetencyAndSkill
  • HCM_JobCredential
  • HCM_JobEducation
  • HCM_JobExperience

See Infor OS User and Administration Library (Cloud) and select User > Data Fabric > Viewing data in Data Lake.

Define the dataset in the AI platform

The AI Platform uses datasets to provide structured input for training and prediction in the position recommendation model. These datasets are derived from replicated HR Talent data that is stored in the Data Lake.

The AI platform combines employee qualifications, job requirements, and position attributes. This combination enables the model to learn relationships between talent profiles and job roles.

Multiple datasets are used during training to represent positions and jobs. These datasets include competencies, credentials, education, and experience. An additional dataset is used as the payload input for real-time prediction.

These datasets ensure that historical patterns and current organizational data are available for accurate model training and inference.

These Data Lake tables are required:

  • HCM_Position
  • HCM_PositionCompetencyAndSkill
  • HCM_PositionCredential
  • HCM_PositionEducation
  • HCM_PositionExperience
  • HCM_Job
  • HCM_JobCompetencyAndSkill
  • HCM_JobCredential
  • HCM_JobEducation
  • HCM_JobExperience
Note: Dataset queries reference the replication table names that are defined in the replication set.

See Infor Artificial Intelligence User Guide (Cloud) and select Content deployment and configuration > Adding the dataset.

Deploy custom algorithm

The custom algorithm defines the machine learning logic for generating position recommendations. Custom algorithms process prepared datasets to train a predictive model that evaluates alignment between employee qualifications and job requirements.

The algorithm applies statistical techniques to identify patterns between skills, competencies, and successful job matches. It uses configurable hyperparameters to influence model behavior and performance.

After deployment, the algorithm is available for use in the training pipeline.

These components are required:

  • Custom algorithm source code package, for example, HCM_CareerPlanning_PositionRecommendation
  • Hyperparameters configuration for model tuning

See Infor Artificial Intelligence User Guide (Cloud) and select Machine learning > Machine learning custom algorithms > Uploading a source file.

Deploy quest and endpoint

The quest represents the end-to-end machine learning pipeline for position recommendations. It orchestrates data preparation, model training, evaluation, and prediction.

The quest connects datasets and the deployed custom algorithm to produce a trained model.

A production quest enables real-time prediction by linking the trained model to an input dataset.

The endpoint exposes this functionality as a callable service. External systems, such as HR Talent, can request recommendation results dynamically.

These components are required:

  • Training datasets
  • Payload dataset
  • Trained model output
  • Predictor configuration
  • Real-time production quest type
  • Endpoint name to use in HR Talent configuration

See Infor Artificial Intelligence User Guide (Cloud) and select Content deployment and configuration > Creating the quest and deploying the endpoint.

Configure and schedule workflow

Workflows automate the retraining and deployment cycle of the recommendation model. They ensure that the model reflects updated organizational data.

Workflows periodically refresh datasets, retrain the model, and redeploy the endpoint.

Scheduling defines how often recommendations are updated. It aligns model refresh cycles with data availability and business needs.

This process helps maintain consistent and up-to-date recommendation results without manual intervention.

These configuration elements are required:

  • Workflow template
  • Service account for workflow execution
  • Schedule configuration for frequency and timing
  • Integration with dataset refresh and model retraining

See Infor Artificial Intelligence User Guide (Cloud) and select Optimization > Model maintenance > Creating a workflow.

See Infor Artificial Intelligence User Guide (Cloud) and select Optimization > Model maintenance > Setting up a workflow schedule.