Résidanat 2025 — Prediction Method (Student Guide)

🧭 Purpose

This page explains how we predict likely courses for Résidanat and how to use the results to study smarter.

This report was prepared by MBset team mbset.com

Join our telegram channel for more https://t.me/mb_set

1) Dataset (What we trained on)

• Scope: >50,000 questions from Externat Alger, Résidanat Alger (more than 20 years of exams).
• Exams: Résidanat Alger (target) + supporting signals from Externat Alger (and others when present).
• Granularity per question:
  • module, cours,
  • year, examType (e.g., Externat or Résidanat)., period (e.g., P1, P2…, or Rattrapage).

2) What we predict

For each cours inside each module, we produce for the next Résidanat:

• Probability of appearance (at least one question).

• Expected number of questions (fractional; later rounded under module constraints).

These feed our one‑click 60‑questions simulation to mirror real blueprints. ONLY ON MBset.com

3) Reason codes (you’ll see these on topics)

• evergreen — appears in ≥80% of years.
• share rising — recent share increasing within module.
• streak k yrs — appeared k years in a row.
• gap k yrs — absent for k years (often due for a return).
• externat preseed (t‑1) — strong in previous‑year externat.