Skip to content

Repository files navigation

African Customer Intelligence Platform

An AI-powered customer analytics platform that transforms raw customer data into actionable business insights through machine learning, interactive dashboards, and automated analytics.

Python Machine Learning Streamlit Power BI Status


Project Team

Project Leader

Thato Maelane

Assistant Leader

Amarachi Florence

Team Members

  • Philip Odiachi
  • Mavis

Developed during the DataVerseAfrica Internship – Cohort 3


Project Overview

The African Customer Intelligence Platform is an AI-powered business analytics solution designed to help organizations quickly uncover valuable insights from customer data.

Instead of spending days creating reports or manually analyzing spreadsheets, users can upload customer datasets and instantly receive interactive dashboards, predictive analytics, customer segmentation, sentiment analysis, geographic insights, and strategic business recommendations.

Although initially developed using banking customer data, the platform is flexible enough to support industries such as:

  • Banking
  • FinTech
  • Telecommunications
  • Retail
  • Insurance
  • Healthcare
  • Customer Service
  • E-commerce

Project Objectives

  • Transform raw customer data into actionable business intelligence.
  • Identify customer behavior and trends.
  • Detect high-risk customer segments.
  • Measure customer sentiment.
  • Analyze digital adoption.
  • Discover high-value customer locations.
  • Support strategic business decision-making.
  • Deliver insights through an easy-to-use web application.

Key Features

Interactive Dashboard

  • Customer overview
  • KPI summaries
  • Business performance indicators
  • Interactive charts

AI-Powered Analytics

  • Customer segmentation
  • Risk analysis
  • Customer profiling
  • Trend detection
  • Predictive analytics

Customer Sentiment Analysis

Analyze customer feedback to identify:

  • Positive sentiment
  • Neutral sentiment
  • Negative sentiment

Helping organizations understand customer satisfaction.


Geographic Intelligence

Visualize customer distribution across different regions to identify:

  • High-value customer locations
  • Regional opportunities
  • Market concentration

Digital Adoption Analysis

Measure customer engagement with digital services including:

  • Mobile application usage
  • Digital banking adoption
  • Customer activity trends

Automated Recommendations

The platform automatically generates business recommendations based on customer behaviour and analytics.

Examples include:

  • Improve customer retention
  • Increase digital engagement
  • Reduce customer churn
  • Target high-value customers
  • Improve customer satisfaction

Example Insights

Using over 5,200 customer records, the platform identified:

  • Approximately 40% of customers were classified as high risk.
  • Only 25% actively used the mobile application.
  • Customer complaints negatively affected overall sentiment.
  • High-value customers were concentrated in major cities such as Lagos and Abuja.

Project Workflow

  1. Upload customer dataset
  2. Data preprocessing
  3. Data validation
  4. Feature engineering
  5. Customer analytics
  6. Machine learning predictions
  7. Visualization generation
  8. Business recommendation engine
  9. Dashboard presentation

Technologies Used

Programming

  • Python

Data Analysis

  • Pandas
  • NumPy

Data Visualization

  • Matplotlib
  • Seaborn
  • Plotly

Machine Learning

  • Scikit-learn
  • XGBoost

Dashboard

  • Streamlit

Business Intelligence

  • Power BI

API Development

  • FastAPI / REST API

Repository Structure

African-Customer-Intelligence-Platform/
│
├── README.md
├── app.py
├── api.py
├── model.pkl
├── customer_dataset.csv
├── presentation.pptx
├── project_report.pdf
├── requirements.txt
├── Images/
└── Models/

Analytics Included

  • Customer Segmentation
  • Customer Risk Analysis
  • Customer Lifetime Insights
  • Geographic Distribution
  • Sentiment Analysis
  • Digital Adoption Analysis
  • Customer Behaviour Analysis
  • Executive Dashboard
  • Automated Business Recommendations

Machine Learning Workflow

  1. Data Collection
  2. Data Cleaning
  3. Exploratory Data Analysis (EDA)
  4. Feature Engineering
  5. Model Development
  6. Model Evaluation
  7. Dashboard Integration
  8. API Development
  9. Streamlit Deployment
  10. Business Intelligence Reporting

Leadership Contribution

As the Project Leader, my responsibilities included:

  • Leading a multidisciplinary team throughout the project lifecycle.
  • Coordinating project planning and task allocation.
  • Designing the machine learning workflow.
  • Guiding data preprocessing and feature engineering.
  • Integrating analytics into the Streamlit application.
  • Ensuring successful collaboration and timely project delivery.
  • Presenting project outcomes and strategic recommendations.

Skills Demonstrated

  • Team Leadership
  • Project Management
  • Machine Learning
  • Predictive Analytics
  • Customer Analytics
  • Business Intelligence
  • Data Visualization
  • Exploratory Data Analysis (EDA)
  • Feature Engineering
  • API Development
  • Dashboard Development
  • Streamlit
  • Power BI
  • Python Programming
  • Model Deployment

Python Libraries

pandas
numpy
matplotlib
seaborn
plotly
scikit-learn
xgboost
streamlit
fastapi
joblib

Future Improvements

  • Real-time customer monitoring.
  • Cloud deployment on AWS or Azure.
  • Integration with SQL databases.
  • Customer churn prediction.
  • Recommendation system.
  • Fraud detection module.
  • Role-based authentication.
  • AI chatbot for business insights.
  • Automated report generation.
  • Multilingual support for African markets.

Live Demo

Streamlit Application

https://african-financial-risk-dashboard-d.streamlit.app/


Power BI Dashboard

https://app.powerbi.com/view?r=eyJrIjoiMDM5MWNiNGUtN2YzYi00YWMxLWE1OWYtM2I5ZjM3YjA0Y2EyIiwidCI6IjA0MjhjMTgzLTc5ZjYtNDJlOC05NmE0LWZiNzIxN2Y2NTg4YiJ9


Project Report

https://docs.google.com/presentation/d/1JN4UVYeWbU-VgZt-d_Dew5ne6vzgucit/edit?slide=id.p1#slide=id.p1


Team

Thato Maelane (Project Leader)

Bachelor of Engineering Technology (Electrical Engineering)

Tshwane University of Technology


Amarachi Florence (Assistant Leader)


Philip Odiachi


Mavis


⭐ If you found this project useful, consider giving the repository a star!

About

AI-powered customer intelligence platform developed during the DataVerseAfrica Internship. Combines machine learning, Streamlit, Power BI, and APIs to transform customer data into actionable insights, predictive analytics, interactive dashboards, and automated business recommendations across banking, retail, telecom, and other industries.

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages