Find your ideal Master's degree
with Artificial Intelligence

"Repartons à zéro avec monmaster-ai. Jouez le rôle d'un statisticien et découvrez quelles formations de master en droit ont attiré le plus de candidats, quelles spécialisations juridiques sont les plus demandées et quelles informations utiles on peut extraire de ces données."

"Using only the MonMaster MCP — no web search, no external rankings — analyse the following formation(s): all Droit des affaires formations at Université Jean Moulin Lyon 3 (search by city Lyon, mention Droit des affaires, establishment Lyon 3) For each formation, retrieve: candidatures confirmées (N), capacité (C), rang du dernier appelé (R), and taux d'accès (A). Then compute and report the following derived metrics: 1. Demand intensity → N / C (applicants per seat) 2. Admission bar → R (absolute rank of last admitted) 3. Yield proxy → C / R (share of offers that convert; low = popular first-choice) 4. Selectivity signal → A = R/N (verify mechanically against raw figures) 5. Offer inflation → R / C (offers made per seat; high = waitlist moves far) Interpret each metric in plain language. Flag any statistical artefact (e.g. micro-cohort distortions, dual M1/M2 entries, apprentissage splits inflating or deflating A%). Produce a probabilistic tiering for a candidate whose likely pool percentile is [[µ — e.g. "top 7%"]] with evaluation noise σ = 4 points, using the formula P(admit) = Φ((A − µ) / σ). Label each programme: HIGH REACH (<15%) · REACH (15–40%) · MATCH (40–80%) · SAFETY (>80%) Run a sensitivity check: recompute P at µ − 3pp and µ + 3pp and state how the tier label changes. Format the output as: • A ranked table (N, C, R, A%, N/C, R/C, yield%, P%, tier) • A prose interpretation of the 2–3 most statistically significant findings • An interactive visual: scatter plot of N (x-axis, demand) vs A% (y-axis, selectivity, inverted) with bubble size = C, colour = tier, and hover tooltips showing all metrics • A reach/match/safety tiering card per programme Do not use any data source other than MonMaster MCP tool calls. After completing the standard analysis above, compute the joint probability of receiving at least one offer at MATCH tier or above across the full portfolio, assuming independent admission decisions: P(≥1 match) = 1 − ∏(1 − Pᵢ) for all programmes where tier ∈ {MATCH, SAFETY} Present this as a single summary statistic and state which 2–3 programmes contribute most to the joint probability (the "load-bearing" choices in the portfolio). "