Customer_Support_Chatbot

Build a “Customer-Support Chatbot” for a bank

Clone source

git clone https://github.com/hungho77/Customer_Support_Chatbot_Zalo.git
cd Customer_Support_Chatbot_Zalo

Installation

Python

Create a conda environment and install the requirements:

conda create -n chatbot python=3.10 -y
conda activate chatbot
pip install -r requirements.txt

Create env

Create file .env (using vim) and setting environment variable

GOOGLE_API_KEY="google api key"
DOCUMENT_PATHS="path to folder contant document as csv files"
QDRANT_HOST="localhost"
QDRANT_PORT="6333"

copy above environment variable into .env file and save

cd src
vi .env

then

cd ../

Run qdrant storage

sudo docker run -d -p 6333:6333 -p 6334:6334 -v $(pwd)/qdrant_storage:/qdrant/storage qdrant/qdrant

Run the application

streamlit run src/app.py

Use the application

open browser with url below
Local URL: http://localhost:8501
Network URL: 

Examples

Download the demo video