CoolFace
Apppublic

RishabhHajela/b2b-customer-segmentation-and-churn-predictor

sourceHugging Faceupdated 3mo agoView on Hugging Face
0likes
App README

B2B Customer Segmentation & Churn Predictor

300 synthetic cloud-services accounts — Large/Medium/Small companies across industries — segmented with KMeans (RFM + utilization + support load) and churn-scored by an interpretable logistic-regression model with named risk drivers per account.

The landing page is an interactive API console — click any endpoint to call the live API.

API

MethodPathDescription
GET/healthHealth check
GET/api/segmentsSegment summary (size, avg CLV, avg churn probability)
GET/api/accounts?segment=X&risk_band=YFiltered accounts
GET/api/accounts/{id}Account profile with churn drivers
GET/api/churn/summaryRisk-band distribution + top at-risk accounts
GET/api/churn/modelAUC / precision / recall + coefficients
POST/api/refreshRecompute segments + churn scores

Stack

Python 3.11 · FastAPI · SQLite · scikit-learn · Pydantic v2 · Next.js 14 (static export) · Tailwind CSS · Docker