Skip to content
View lebede-ngartera's full-sized avatar

Highlights

  • Pro

Block or report lebede-ngartera

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
lebede-ngartera/README.md

Lebede Ngartera

AI Strategist & Machine Learning Engineer
PhD in Applied Mathematics · Bayesian & Probabilistic Machine Learning
TeraSystemsAI · Research in Healthcare, Security, and Industrial AI

I design and build reproducible, uncertainty-aware machine learning systems for real-world decision-making, with particular focus on high-stakes domains such as fraud detection, customer retention, healthcare, and industrial engineering.

My work connects peer-reviewed research with production-grade ML systems, emphasizing interpretability, evaluation rigor, and risk-aware modeling.

Selected Research Publications

Title Journal Year DOI
Stochastic Inventory Optimization with Coherent Risk Measures: A Decision-Theoretic Framework for Probabilistic Forecasting and Constrained Optimization Journal of Risk and Financial Management 2026 https://doi.org/10.3390/jrfm19030173
Bayesian RAG: Uncertainty-Aware Retrieval for Reliable Financial Question Answering Frontiers in Artificial Intelligence 2026 https://doi.org/10.3389/frai.2025.1668172
Hybrid Naïve Bayes Models for Scam Detection IEEE Access 2025 https://doi.org/10.1109/access.2025.3569216
Enhancing Autonomous Systems with Bayesian Neural Networks Frontiers in Built Environment 2025 https://doi.org/10.3389/fbuil.2025.1597255
Application of Bayesian Neural Networks in Healthcare Machine Learning and Knowledge Extraction 2024 https://doi.org/10.3390/make6040127

Applied Machine Learning Projects

Customer Churn Risk Prediction

End-to-end machine learning pipeline for customer churn prediction using real-world tabular data, with a focus on reproducibility, leakage control, evaluation discipline, and explainability.

Repository: https://github.com/lebede-ngartera/customer-churn-risk-ml


Decision Intelligence & Operations Research

Supply Chain Decision Intelligence

Decision-intelligence project integrating probabilistic demand forecasting, constrained optimization, and Monte Carlo risk evaluation for supply chain planning under uncertainty.

The project shows how uncertainty-aware forecasts ($P50$ vs $P90$) lead to materially different operational decisions under inventory, capacity, and budget constraints, and how CVaR-based stress testing exposes tail risk that mean-based metrics miss.

Focus areas:

  • Forecast-to-decision linkage
  • Optimization under operational constraints
  • Risk-aware evaluation via Monte Carlo simulation
  • Executive-style decision memos communicating cost, service, and risk tradeoffs

Repository: https://github.com/lebede-ngartera/supply-chain-decision-intelligence


Industrial AI & Multimodal Systems

GeoFusion AI

Industrial multimodal AI platform for 3D CAD geometry understanding, combining point-cloud learning, graph neural networks, retrieval, anomaly detection, uncertainty-aware prediction, and generative modeling.

The project unifies 3D geometry, text, and engineering metadata into a shared embedding space to support similarity search, anomaly detection, text-to-shape retrieval, and engineering decision support in CAD/CAE workflows.

Focus areas:

  • Multimodal representation learning for industrial data
  • 3D geometric deep learning with PointNet++, DGCNN, and GNNs
  • Retrieval and anomaly detection for engineering workflows
  • Uncertainty-aware property prediction and risk-sensitive modeling
  • Interactive ML system design for technical users

Repository: https://github.com/lebede-ngartera/GeoFusion-AI


Scam Detection via Hybrid Bayesian Models

Implementation and evaluation of hybrid Naive Bayes-based models for real-world scam detection, derived from peer-reviewed research.

Focus areas:

  • Bayesian generative modeling under data sparsity
  • Cost-sensitive evaluation in highly imbalanced settings
  • Uncertainty-aware decision thresholds

Derived from: Hybrid Naïve Bayes Models for Scam Detection (IEEE Access, 2025)
Repository: (in progress)


Research → System Philosophy

  • Research defines what is theoretically sound
  • Engineering determines what is deployable
  • Evaluation decides what is trustworthy

I focus on making tradeoffs explicit and uncertainty visible rather than optimizing single-point metrics in isolation.


Areas of Interest

  • Fraud and scam detection
  • Customer behavior modeling
  • Bayesian and probabilistic machine learning
  • Risk-aware AI systems
  • Industrial AI for engineering workflows
  • Interpretable ML in regulated environments

Pinned Loading

  1. customer-churn-risk-ml customer-churn-risk-ml Public

    End-to-end machine learning pipeline for customer churn risk prediction using real-world tabular data, with a focus on reproducibility, evaluation, and explainability.

    Python

  2. lebede-ngartera.github.io lebede-ngartera.github.io Public

    Portfolio

    HTML

  3. orcid orcid Public

    🔬 ORCID Portfolio - Professional academic showcase featuring research publications, ORCID credentials, and AI expertise. Part of the Golden Edge Technology Showcase series.

  4. portfolio portfolio Public

    GitHub Profile & Portfolio

    HTML