Open-source Jev-style System One decision model. Gemma 3 270M with a scoring head — fast, calibrated decisions in a single forward pass. No text generation. Inspired by TypeSafe.ai's Jev.
EDA & probability analysis of airline baggage delivery times across ~99K real flights. Python · pandas · SciPy · Matplotlib.
A humorous, probability-based web app that calculates humanities graduate's odds of surviving a data analytics course. Features an independent-factor model. Project materials & code in Russian.
Excel-based weighted resource allocation model with normalized scoring and proportional reward distribution.
达人/买手四维评估模型：可解释打分、分层与差异化动作建议
Python decision engine for scoring, ranking, and standardizing multi-factor decisions using configurable inputs and weights.
Private-network (PMR) market intelligence POC: archive public vendor data, build a 5-dimension capability matrix, score vendor fit against a bid scenario, output a reproducible HTML report. 专网通信市场情报与竞品分析 POC：公开信息归档 → 五维能力矩阵 → 场景化匹配度评分 → 可复现报告。
A Python tool for scoring and prioritizing procurement opportunities using margin, technical fit, delivery feasibility and risk.
可解释的八维人才评估：可配置加权评分 + S+/S/A/B/C 分层 + NOT EXISTS 排他 AB 实验（模拟数据）
Open-source KYB (Know Your Business) verification platform for fintech, merchant onboarding, compliance, AML, and business due diligence. Automates company verification, risk scoring, data collection, and report generation with AI-powered background agents.
Which companies does AI actually disrupt? A 116-column company schema, an evidence-scored exposure model (EPI), LLM research agents that cite sources, and an interactive constellation dashboard.
Credit Scoring Model using Logistic Regression, Decision Tree and Random Forest on the German Credit Dataset.
Analysis for a jewelry repair workshop in Saint Petersburg – two-stage scoring model using open demographic and competitive data
Interactive Streamlit app that converts Politiscales-style scores into political profiles, x/y spectrum coordinates, detailed interpretations, and comparisons against a 500-profile reference dataset.
Weekly FPL player scoring model — ranks all 500+ players per position using a season-aware blend of last-season PPG, current form, fixture difficulty, and position-specific stats. Surfaces top picks, differentials, and watchlist players.
To associate your repository with the scoring-model topic, visit your repo's landing page and select "manage topics."