Skip to content

A sentiment analysis project driven by real time data collection, a combination of CNN-LSTM architecture that performs sentiment analysis on customer reviews and the Gemini LLM API that provides relevant marketing strategies with a well-designed and user-friendly interface

Notifications You must be signed in to change notification settings

rahulsm20/marketing-sentiment-analysis

Repository files navigation

Marketing Strategy Optimization using Sentiment Analysis

Introduction

A sentiment analysis project driven by real time data collection and a combination of CNN-LSTM architecture.

Index

Prerequisites

Setup

Setup Notebook

  • Create virtual environment
python -m venv venv
  • Enter virtual environment

    • Bash

      source venv/bin/activate
      
    • Powershell

      .\venv\Scripts\activate
      
  • Install requirements

pip install -r requirements.txt

Setup

Scraping Service

cd scraping-service
  • Install modules

    npm install
    

    or

    npm i
    
  • Setup environment variables

    $ scraping-service/.env

    MONGO_URL=
    API_KEY=
    POSTGRES_URL=
    
  • Starting the servers
    $ scraping-service
    npm run dev
    

Generation Service

cd generation-service
  • Install Packages

    pip -r requirements.txt
    

$ generation-service/.env

GEMINI_API_KEY=

$ generation-service

uvicorn main:app --host 0.0.0.0 --port 8000

Setup Client

  • Setup environment variables
    VITE_AUTH0_CLIENT_ID=
    VITE_AUTH0_DOMAIN=
    VITE_SCRAPING_SERVICE_URL=https://localhost:5000
    VITE_GENERATION_SERVICE_URL=https://localhost:8000
    VITE_CLIENT_URL= https://localhost:5173/
    VITE_SERVER_URL=https://localhost:5000
    
    $ client
    npm i && npm run dev
    

Using Docker

$ .

docker compose up

System Design

system

About

A sentiment analysis project driven by real time data collection, a combination of CNN-LSTM architecture that performs sentiment analysis on customer reviews and the Gemini LLM API that provides relevant marketing strategies with a well-designed and user-friendly interface

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published