units4u

👍🎭👎 a university unit recommender system based on natural language (sentiment)

for the second sprint of our web research unit, you can read our paper or go look at the demo site (note it's on heroku so may take some time to boot)

install

most of this project will be using node so you'll need to install any dependencies. we're using eslint to lint the project.

make sure you're using a node version 4 or above for es6 (es2015) features

1. npm install -g eslint
2. npm install
3. Install and start neo4j
4. copy the .env-sample file to .env and add details (username and password for neo4j, indeed api publisher api key).
5. [Optional] node app/test - Test it's working

neo4j

you'll need to download and install neo4j (community edition).

to drop the database you can use the cipher command: MATCH (n) DETACH DELETE n

run

1. neo4j start - start neo4j
2. npm start - will seed db automatically if not seeded

code styleguide

we're using EditorConfig to set-up our coding style, you can install a plugin for this to take effect

e.g for atom apm install editorconfig

our linter will also pick up things like this as well

linting

linting is done with eslint, you can install it your ide. running npm run lint will run the linter command line.

deployment

make sure you build the latest clientside script to ES5 using npm run build, add it to the repo for now and push up

/experiments

section for experiment with our different tech and algorithms

/resources

currently we're using this folder to dump data that'll we'll process in the future

credit

using the wonderful twemoji emoji set