🥗 Silent Hunger Discovery Engine (SHDE)

Predictive Analytics for Micronutrient Deficiency Risk in South Asia (2010–2030)

---

📌 Project Overview

"Silent Hunger" (micronutrient deficiency) affects over 2 billion people globally. Unlike visible malnutrition, it often remains undetected until irreversible health damage occurs.

The Silent Hunger Discovery Engine (SHDE) is a research-driven predictive analytics project that integrates:

• Public health indicators
• Economic stress signals
• Nutritional trend data

This study focuses on Pakistan and India (2010–2025) to examine how economic variables—particularly Food CPI (Food Inflation)—influence national-level nutritional outcomes.

---

🔍 Key Research Questions

• Temporal Trends:
  How have anemia, stunting, and wasting prevalence evolved over the last 15 years?

• Economic Correlation:
  Does higher Food CPI (inflation) correlate with increased micronutrient deficiency risk?

• The Shadow Effect:
  How does food inflation today affect nutritional health 12 months later?

• Predictive Analytics:
  Can we forecast micronutrient deficiency risk for 2030 based on current trajectories?

---

🛠 Methodology & PACE Roadmap

Phase	Notebook	Objective	Key Outcome
Plan	01_Research_Mapping	Define theoretical framework & hypotheses	Structured indicator framework
Analyze	02_Data_Acquisition	Ingest & harmonize multi-source global datasets	Wide-to-Long (Melted) dataset
Construct	03_EDA_Statistics	Pattern discovery, outlier handling, feature engineering	Clean Master Dataset
Execute	04_Modeling_Forecast	Machine Learning modeling & risk scoring	74.4% Accurate Random Forest Model + 2030 Forecast

---

📈 Key Findings

• 12-Month Lag Effect:
  Inflation impacts households immediately, but peak nutritional deterioration appears approximately one year later.

• Country Sensitivity:
  Pakistan shows higher nutritional risk sensitivity to Food CPI volatility compared to India.

• Feature Importance:

  • Food CPI
  • Country-specific economic baselines
  • Inflation-adjusted indicators

  These emerged as the strongest predictors of micronutrient deficiency risk.

---

💻 Technology Stack

• Language: Python 3.x
• Core Libraries: Pandas, NumPy, SciPy, Scikit-Learn
• Visualization: Seaborn, Matplotlib, Plotly
• Framework: PACE (Plan – Analyze – Construct – Execute)

---

🎯 Scope & Ethics

• Level of Analysis: Population-level (National/Regional)
• Non-Clinical Use: Designed for research and policy insights — not for individual diagnosis
• Reproducibility: Fully documented preprocessing steps and model assumptions

---

📂 Repository Structure

├── Notebook_01_Research_Problem_Mapping.ipynb

├── Notebook_02_Data_Acquisition_Structure.ipynb

├── Notebook_03_EDA_Statistical_Discovery.ipynb

├── Notebook_04_Modeling_Forecasting_Insights.ipynb

├── SHDE_Forecast_2026_2030.pdf

├── shde_master_engine_v1.xls

└── README.md

---

🤝 Connect & Collaborate

I am a Data Scientist focused on social-impact analytics and public health modeling.

Shanzay Khan