ICSPE — Customer Segmentation Engine
Class Demo — Live

Turn raw transactions
into customer intelligence

An end-to-end analytical engine that ingests e-commerce data, computes RFM behavioral metrics, segments customers via K-Means clustering, and predicts churn — all in-browser, no backend required.

Jump to
1
Data
2
Segments
3
Predict
4
Dashboard
System pipeline — click any step
01
Data Ingestion
CSV upload, schema validation, integrity checks
Open →
02
Feature Engineering
Compute RFM — Recency, Frequency, Monetary
Open →
03
Segmentation
K-Means clustering by behavioral similarity
Open →
04
Prediction
Churn probability and CLV estimation
Open →
05
Reporting
Interactive dashboards and recommendations
Open →
⬡
Architecture
4-Tier Layered
Data · Analytics · App · Presentation
◈
Design Patterns
Repository · Strategy
Observer · Facade
⟳
ML Algorithms
K-Means + Logistic
Segmentation + churn prediction
▤
Sample Dataset
200 Transactions
80 customers · 6 columns

Data Ingestion

Upload transactional data — validation runs automatically before processing

Layer 1 — Data
1
Upload
2
Validate
3
Process
4
Ready
📂

Click to load sample dataset

Pre-built 200-record dataset — one click for a full live demo

sample_transactions.csv 200 records 6 columns

Customer Segmentation

K-Means clustering applied to normalized RFM feature matrix

Layer 2 — Analytics
◎

Load transaction data to run segmentation

Churn Prediction & CLV

Logistic regression — churn probability and customer lifetime value estimates

Layer 2 — Analytics
◎

Load data and run segmentation to generate predictions

Executive Dashboard

Aggregated insights and recommendations for marketing and CRM teams

Layer 4 — Presentation
◎

Load transaction data to populate the dashboard

Done