AI Model Evaluation and Predictive Modeling – Digital Twin Framework (Master’s Thesis)
Text
Developed, applied, and refined LSTM neural networks for predictive modeling as part of a digital twin framework for vessel motion monitoring. Assessed model outputs by validating simulation results against ground truth sensor data, providing feedback to improve prediction accuracy. Produced technical documentation detailing model evaluation processes and outcomes.
• Evaluated AI-generated outputs in engineering contexts for accuracy and robustness.
• Conducted model architecture tuning and hyperparameter optimization based on assessment results.
• Summarized findings for technical documentation and reporting.
• Engaged in iterative model refinement using feedback from evaluations.
Developed, applied, and refined LSTM neural networks for predictive modeling as part of a digital twin framework for vessel motion monitoring. Assessed model outputs by validating simulation results against ground truth sensor data, providing feedback to improve prediction accuracy. Produced technical documentation detailing model evaluation processes and outcomes. • Evaluated AI-generated outputs in engineering contexts for accuracy and robustness. • Conducted model architecture tuning and hyperparameter optimization based on assessment results. • Summarized findings for technical documentation and reporting. • Engaged in iterative model refinement using feedback from evaluations.
2025 - 2026