BeyondStats

Tech Venture · Data + Machine Learning

Gender inequality, in one number you can actually read.

Inequality data is scattered across dozens of disconnected indicators and dense reports. BeyondStats combines them into a single, interpretable score — uses machine learning to estimate the years official data is missing — and makes the whole thing explorable through interactive visuals.

Inequality score explorer
composite GII · 0 = parity, 1 = max inequality

Contributing dimensions

Economic
Social
Physical / Health
GII trend 2018 → 2024 (★ = ML-predicted)

↑ A scripted illustration with representative values — pick a country and year to explore the composite score and its dimensions.

01 — The problem

Too many indicators, too little understanding.

Gender inequality is usually presented through dozens of disconnected metrics and complex reports, which makes it hard to see patterns, compare countries, or engage meaningfully. BeyondStats lets people focus on understanding inequality instead of decoding numbers.

🧮

One score

Multiple gender-related metrics combined into a single, interpretable number.

🤖

ML fills the gaps

Machine learning estimates recent values where official indices are missing.

🗺️

Visual, not raw

Insights through interactive maps and trends instead of raw tables.

02 — The machine-learning score

A Random Forest that reads inequality.

Random Forest regression → GII

Trained on real historical Gender Inequality Index values, the model predicts GII for 2022, 2023, and 2024 — the years where no official data was available — turning dozens of indicators into one clear score.

EconomicSocialPhysical / Health

Input features are grouped into three categories — economic, social, and physical/health. Without ML, users would have to manually compare dozens of indicators; the model reduces that to one number, making comparison and exploration intuitive.

03 — Key features

Explore, don't decode.

🌍

Interactive world map

Country-level exploration at a glance.

📈

Country insights

Historical trends per country over time.

🎯

ML inequality score

The composite, machine-learning-powered score front and centre.

🕹️

Mini-games

Learning through interaction, not just reading.

📊

Three dimensions

Economic, social, and physical/health factors, broken out.

🔎

Compare countries

Put places side by side on one consistent scale.

04 — Stack & data

Built with.

⚛️ React + TypeScript ⚡ Vite 🌲 Random Forest regression 🐍 Python (scikit-learn) 🗺 Interactive maps 📂 World Bank Gender Statistics