Skip to content

davidnene/tunga-AKI-hackathon

Repository files navigation

AKI Insurance Hackathon Project - AI Powered Software by Tunga Team

Overview

This repository contains the source code for the AI-driven software developed during the AKI Insurance Hackathon. The project leverages a Jupyter notebook analysis, Insights Dashboard and a Chatbot to provide a user-friendly interface where stakeholders can interact directly with data insights and make informed decisions quickly.

Features

  • Data Analysis Notebook: Contains comprehensive data cleaning and analysis of insurance data including customer demographics, policy details, and claims history.
  • Dashboard: A dashboard containing a report of insights generated from the insurance dataset.
  • Chatbot: A chatbot that allows users to query the dataset interactively through a chat interface.
  • AI-Driven Insights: Utilizes Chat GPT-4 to analyze data and generate actionable insights, enhancing decision-making processes.

Live Link

https://tunga-chat-hackathon.streamlit.app/

Reports Page

Reports

Reports

Chatbot Page

Home

Data Upload

data Upload

Text QA

Text QA

Visualization Generation QA

Chat Visualization Generation QA

Installation

Prerequisites

  • Python 3.8+
  • pip

Setup (Using bash or Unix terminal)

  1. Clone the repository:
    git clone https://github.com/davidnene/tinga-AKI-hackathon.git
    cd tunga-AKI-hackathon
  2. Create a virtual environment:
    python -m venv .venv
    
  3. Activate the virtual environment
    . .venv/Scripts/activate
  4. Install Dependancies from requirements.txt
    pip install -r requirements.txt
  5. Create an environment file
    touch .env
  6. Add your Open AI API Key
    echo OPENAI_API_KEY=paste_your_key_here > .env
  7. Run streamlit server
    streamlit run intro.py
    

Project Contributors

This project was developed by Strathmore University students pursuing a Masters Degree in Data Science and Analytics

  1. David Nene - Software Engineer|AI
  2. Derrick Lubanga - Data Scientist
  3. Linda Kelda - Data Engineer
  4. Chepkirui TOnui - BI Analyst

Presentation Date

10th-May-2024

About

Large Language Models Innovation in Insurance

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages