Fraud Detector App

An app that help Fraud Officers see transactions and figure out if it is fraudulent or not, depending upon many different parameters.

This app was boostrapped using a simple starter with frontend using Angular 4+ and API server with NodeJS and Express framework. Typescript is configured for both Frontend and API server.

Frontend is created using angular-cli. So we can use all of the angular-cli commands from client folder.

This app can be used in two different ways:

1. Separate Frontend and API server.

2. Both Frontend and API served through single node server in production.

Application Screenshot

Technology

• NodeJS

• Angualr

• MongoDB

• Typescript

Installation

• clone this repository

• npm run app-install to install all dependences

• Ensure you have Mongo installed, if not download and install here

Docker

• Pull image from docker hub by running docker pull melmichael/fraud-detector-app

• Run the docker image. E.g. docker run -p 4200:4200 -p 5000:5000 melmichael/fraud-detector-app

Commands

• Run mongod to get the mongo instance up and open a way for database connection.

• npm run start: Starts both frontend and API server as separate live reload server. Frontend uses ng serve underneath and API server uses nodemon for live reload.

• npm run build: Creates dist folder in both client and server folders.

• npm run start:prod: Builds the client and server projects and starts it with NODE_ENV=production. This by default sets express.static to client/dist to use this node server to serve frontend also.

• Visit localhost:4200 to view the client side and localhost:5000 for the server side.