Merchant-KYC-analysis

🧾 Merchant KYC Analysis

A full-stack compliance analytics solution designed to evaluate merchant onboarding, KYC verification, and risk profiling. This project analyzes 100,000+ merchant records to empower financial institutions and compliance teams with predictive insights and interactive dashboards for fraud prevention and regulatory adherence.

🚗 GitHub Project Repository

🔗 Click to view Merchant-KYC-analysis

🧠 Project Overview

KYC compliance is a cornerstone of financial integrity and fraud mitigation. This project delivers an end-to-end analytics platform that enables:

📊 Verification tracking and document validation
🛡️ Risk profiling and review prediction
📅 Onboarding timeline analysis
📈 Executive dashboards for compliance monitoring

🎯 Key Objectives

Clean and preprocess merchant KYC data
Engineer features for review status and risk modeling
Build classification models to predict compliance outcomes
Deploy interactive dashboards for stakeholder decision-making

📁 Project Structure

File Name	Description
`merchant_kyc_100000.csv`	Raw dataset with merchant KYC records
`cleaned_merchant_kyc.csv`	Preprocessed dataset with feature engineering
`kyc_review_model.pkl`	Trained model for predicting review status
`merchant_kyc.sql`	SQL queries for data extraction and filtering
`sqlconnect.py`	Python script for SQL database connection
`app.py`	Streamlit app for dashboard deployment
`merchant_kyc.ipynb`	Jupyter notebook with EDA, modeling, and insights
`merchant_kyc_dashboard`	Power BI or Streamlit dashboard visualizing compliance metrics

🧹 Data Preprocessing

Verified PAN and GST formats
Encoded categorical features (address_proof_type, kyc_status, risk_level)
Converted onboarding_date to datetime format
Normalized compliance_score
Removed duplicates and ensured type consistency

📈 Exploratory Data Analysis

📊 Distribution of KYC status across address proof types
🛡️ Compliance score trends by risk level
🔍 Review status breakdown by document validity
📅 Onboarding timeline and volume analysis
🧠 Correlation matrix of compliance features

🤖 Modeling Approach

Target Variable: review_status
Algorithms Used: Logistic Regression, Random Forest, XGBoost
Evaluation Metrics: Accuracy, Precision, Recall, F1 Score
Top Features: compliance_score, kyc_status, address_proof_valid, risk_level