Linked.Archi ML-Enabled Systems Taxonomy

ML-Enabled Systems Concept Scheme

• ▼By Lifecycle Phase
  • Data Preparation
    Phase covering data collection, cleaning, labeling, and feature engineering.

    See Also: Dataset

    https://meta.linked.archi/ml-systems/tax#DataPreparation
  • Model Development
    Phase covering algorithm selection, training, evaluation, and experiment tracking.

    See Also: MLModel

    https://meta.linked.archi/ml-systems/tax#ModelDevelopment
  • Model Deployment
    Phase covering model packaging, serving infrastructure setup, and production release.

    See Also: ServingInfrastructure

    https://meta.linked.archi/ml-systems/tax#ModelDeployment
  • Model Monitoring
    Phase covering production monitoring, drift detection, and retraining triggers.

    See Also: MonitoringComponent

    https://meta.linked.archi/ml-systems/tax#ModelMonitoring
• ▼By Component Role
  • Inference Component
    Components that perform prediction or inference — models and serving infrastructure.

    See Also: MLModel

    https://meta.linked.archi/ml-systems/tax#InferenceComponent
  • Data Management Component
    Components that manage data flow — pipelines, feature stores, datasets.

    See Also: DataPipeline

    https://meta.linked.archi/ml-systems/tax#DataManagementComponent
  • Orchestration Component
    Components that orchestrate ML workflows — training pipelines, experiment trackers.

    See Also: TrainingPipeline

    https://meta.linked.archi/ml-systems/tax#OrchestrationComponent
  • Governance Component
    Components that support ML governance — model registries, monitoring, audit trails.

    See Also: ModelRegistry

    https://meta.linked.archi/ml-systems/tax#GovernanceComponent
• ▼By Stakeholder Domain
  • Analytics Modeling (Data Science)
    The domain of data scientists — algorithm selection, model training, evaluation.

    https://meta.linked.archi/ml-systems/tax#AnalyticsModeling
  • Analytics Operations (Data Engineering)
    The domain of data engineers — deployment, serving, monitoring, pipeline management.

    https://meta.linked.archi/ml-systems/tax#AnalyticsOperations
  • ML Governance
    The domain of ethics officers and compliance — fairness, explainability, privacy, audit.

    https://meta.linked.archi/ml-systems/tax#MLGovernance