/product-review-OpenAI-summary

A customer feedback demo with OpenAI summarization

Primary LanguageTypeScriptMIT LicenseMIT

Product Review and management demo app

Introduction

A customer review demo application to collect product reviews after a successful purchase with an OpenAI GPT-3 analytics page to analyze natural language reviews.

CircleCI

Table of contents

Major Features

  • View all comments/feedback/reviews from customers on a product.
  • Customer can create feedback by entering their name, email, rating and comment
  • Display product performance based on the reviews by showing a graphical preview in a chart format and the percentage of reviews above 2.

Upcoming features

  • Paginate records for reviews.
  • Application logging with tools like Sentry and Hotjar
  • Husky for running pre-commit tests before pushing code to SCM's like GitHub

Preview

This applicaiton is currently deployed on vercel, a demo preview can be seen on this Preview Link

App Preview

Tools and technologies

Below are the main tools and technologies used in this project:

Getting Started

  1. Setup environment variables

    • Create a .env file at the root directory of the project
    • Copy the content of the .env.example file onto it
    • Update the values as required. e.g. API_URL=http://localhost:3000/api

  2. Development and Testing

    Notes

    Before proceeding with this section, kindly ensure that your terminal window is open on the root directory of the project, you can use the below helper commands to check/confirm:

    • cd command to change directory e.g. cd path/to/project
    • pwd command to check your "print working directory"
    • ls command to show the content of your current directory

    By default the dev script runs on port 3000, if you have another application making use of that port on your machine, you can either terminate that process/application or use the --port flag where necessary to change the running port, kindly remember to update any relative script found on the project e.g. the package.json and Dockerfile.

  3. Create OpenAI tenant in Microsoft Azure (for openai page)

  • Create Azure OpenAI tenant step by step
  • Deploy the model text-davinci-003

Running locally

# Step 1: Install dependencies
$ npm install

# Step 2: Run Nextjs and cypress in interactive mode concurrently
$ npm run develop

Extra notes

Other commands can be found on the package.json file in the root directory of the project.

Viewing test coverage

Please note that you need to have run the cypress tests at least once before a report can be generated which would be located here: coverage/lcov-report/index.html, you can directly open it by navigating to the folder or via your terminal with the below command:

$ open coverage/lcov-report/index.html

Using Docker

Before proceeding with the below command, ensure Docker is installed and running on your machine.

# Runs Nextjs within a docker container
$ docker-compose up

Author

Precious OSSAI

Inspiration

Disclaimer

This is a demo project for proof of concept only.