Luffy-2520/infotact-project1-grievance-nlp

An AI-powered system to automatically analyze citizen complaints: it predicts the responsible government department, classifies sentiment (Critical/Negative/Neutral/Positive), assigns an urgency score (0–100) and recommends priority actions. Intended for municipal grievance handlers, civic-tech teams, researchers or prototype evaluators working on automating complaint triage. Key features and implementation: - Models: TF-IDF features, Random Forest for department classification (53.33% accuracy), SVM for sentiment (66.11%), overall macro F1 53.33% - Stack: Python, NLTK, scikit-learn, FastAPI (REST endpoints: GET /, GET /health, POST /analyze) - Provides urgency scoring and a deployable API; project developed over a four-week internship with daily commits.

このRepoのdescription / GitHub topics / READMEのAI要約に出現したAI関連キーワードをカテゴリ別に表示。各バッジは該当カテゴリの詳細ランキングへリンクします。