DESCRIPTION
-----------

PromotEd
--------
PromotEd recommends courses from multiple MOOC providers based on the skills required for a specific job role. The skills required for each job is collected beforehand from available datasets. Similarly, information about available courses are also procured from multiple MOOC platforms like Udacity, Udemy, and edX. At runtime, all these datasets are used to recommend courses to the user.



Usage
-----
Go to https://promoted.herokuapp.com to view the app.

Since we have a limited number of requests in the free tier per month, this may not work. If so, please follow the steps below to setup the app on your system.



ALTERNATIVE: INSTALLATION
-------------------------
After retrieving the GitHub code, perform the following steps to setup the application.



Running the server:

cd to the CODE directory and install dependencies by executing pip install -r requirements.txt.
Start the server with the following command:

python server.py

Flask should be installed and Python v3.x should be used. The command may change to python3 server.py in some systems.

Running the client (user interface):
cd to the CODE/ui directory and start the interface with yarn.
yarn install
yarn start

Alternatively, npm can also be used instead of yarn.
npm install
npm start



EXECUTION
---------
Go to https://promoted.herokuapp.com to view the app.

Since we have a limited number of requests in the free tier per month, this may not work. If so, please follow the steps in the Installation section to setup the app on your system.

Once you have the client and server running (follow steps above), open your browser and go to http://localhost:3000. This opens up the home page of the application.

Datasets
--------
(a) Job skills dataset:

We used two datasets for this application.

* https://www.kaggle.com/PromptCloudHQ/us-jobs-on-monstercom
* https://www.kaggle.com/PromptCloudHQ/usbased-jobs-from-dicecom

We did some custom transformations on the datasets using OpenRefine. Some of the transformations include

* Clustering similar job titles
* Standardizing vocabulary e.g. converting Software Engineer, Software Development Engineer etc., to Software Developer

The cleaned and transformed datasets are provided under /data/jobs-data in the files dice_US_jobs-clean.csv and monster_com_jobs-clean.csv

The job skillset keywords are extracted by executing keywords.py provided under /data/jobs-data.

(b) Course catalog dataset:

We used the data from three platforms for this project - Udemy, Udacity and edX. The details about the APIs and the scripts used are provided in respective folders under /data.

Some of our transformations were done through OpenRefine, and also, the datasets are in the order of MB. So we have included them. This is to ensure that the working demo can be run locally.

APIs
----
> GET /get_courses?job_title=_JOB TITLE_

The response is an array of JSON objects each of which represents a course. An example is given below.

  [
    {
        "duration": 1,
        "duration_unit": "week",
        "level": 1,
        "price": 149.99,
        "site": "udemy",
        "title": "R Programming: Mastering R for data analysis"
    },
    {
        "duration": 1,
        "duration_unit": "week",
        "level": 1,
        "price": 29.99,
        "site": "udemy",
        "title": "Certified business analysis professional (CBAP) practice"
    }
  ]

> GET /get_jobs

The response is an array of job titles.

[
  "Software Developer",
  "Senior Mobile Developer"
]