Skip to content

eisoptrophobicc/data-intel

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

36 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Conversational BI Backend

AI-powered backend that converts natural language queries into executable SQL, enabling users to analyze datasets without writing queries manually.

Built during a hackathon — Top 2 project


Overview

This system processes user-uploaded datasets and allows querying using plain English.

It is designed to be dataset-agnostic, dynamically adapting to different data structures without requiring predefined schemas.

It dynamically:

  • Interprets intent using LLMs
  • Converts intent → SQL queries
  • Executes queries on structured data
  • Returns analytical results

Key Features

  • Natural language → SQL query conversion
  • Automatic dataset ingestion (CSV → SQLite)
  • Dataset-agnostic querying (works across arbitrary datasets)
  • Modular backend design for flexible query handling
  • LLM-assisted data retrieval pipeline
  • Support for dynamic analytical queries

Architecture

User Query

Intent Extraction (LLM)

SQL Generation

Database Execution

Result Output


Setup

Clone the repository

git clone <repo-url>
cd HACK-A-DON/backend

Install dependencies

pip install pandas requests chardet numpy

Usage

Simply upload a dataset (CSV), and the system will automatically:

  • Process and clean the data
  • Convert it into a database
  • Enable natural language querying

Run:

python run_query.py

Example queries:

Average views in US
Likes by region
Top performing categories

Notes

  • CSV data is automatically processed and stored internally
  • Database setup is handled dynamically
  • Dataset-agnostic design enables reuse across different datasets
  • Developed as a hackathon prototype

About

Conversational AI system that converts natural language queries into SQL and generates business intelligence insights from structured data.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors