Testing SalesForce Einstein on REI Search

This is a demo project to show how a website's SEARCH feature can take advantage of Eintein Intent and make educated decision to route users to the right resources.

https://rei-einstein-intent.herokuapp.com/

- git
- NPM

Intall git from: https://git-scm.com/downloads

OR

Install git through Homebrew:

$brew install git

Intall NPM from: https://www.npmjs.com/get-npm

$npm install

$grunt build

$npm start

• GitHub: Project Repository
• NPM: Used to package the project
  • ExpressJS: Web Framework
  • Request: to make http calls
  • Joi: validator for JavaScript objects
• Grunt: JavaScript Task Runner
  • grunt-contrib-jshint: detect errors and potential problems in your JavaScript code
  • grunt-contrib-sass: compile SASS files to css
  • grunt-contrib-cssmin: minify css files
  • grunt-contrib-uglify: minify js files
  • grunt-contrib-watch: run predefined tasks whenever watched file patterns are added, changed or deleted

• / : root directory. contains most of the config files
  • /bin : contains startup scripts
  • /data : data used to train Einstein
  • /lib : contains custom middleware libraries for the project
  • /public : contains publicly accessable files /public/dist : created using grunt task. contains minified js & css files for production env
  • /routes : contains files that handles the website routes
  • /views : contains the view templates

Since this project is publicly aviable this project does not contain the private key nor the functions to create the token to access the Einstein Intent API.

If the token has experied, please email me for a new one OR one can use their their own account and train Einstein Intent using the data found in data directory. And update the token and modelId variables located in /lib/einsteinIntent.js, with their own data.

1. Create Developer's Account @ https://developer.salesforce.com/einstein
2. Make sure you save the Private Key once you create the account
3. Use your Private Key & Email to create Token @ https://api.einstein.ai/token
4. Upload the data to Einstein Server. datasetId will returned on a successful upload
   • curl -X POST -H "Authorization: Bearer " -H "Cache-Control: no-cache" -H "Content-Type: multipart/form-data" -F "data=PATH_TO\data\training-data.csv" -F "type=text-intent" https://api.einstein.ai/v2/language/datasets/upload
5. Train the dataset. On a successful submission you will get a modelId
   • curl -X POST -H "Authorization: Bearer " -H "Cache-Control: no-cache" -H "Content-Type: multipart/form-data" -F "name=Weather Intent Model" -F "datasetId=YOUR-DATASET-ID" https://api.einstein.ai/v2/language/train
6. Update config in /lib/einsteinIntent.js with your token and modelId

API details can be found @ https://metamind.readme.io/docs/