/College-Recommendatoin-App-MSiA-423

College Application Recommentaion

Primary LanguageJavaScript

College Recommendation Flask App

Project Objective

This repo can be used to produce a U.S. college recommendation web app. The data preprocessing steps are written with R and the app is written with Python 3. The app is currently running at http://ec2-52-91-59-235.compute-1.amazonaws.com:5000/home.

Team Members

  • Developer: Junxiong Liu
  • Product Owner: Zili Li
  • QA: Chris Rozolis

Project Charter

Create a web app to help high school students and parents make well-informed decisions in the college application process based on their preference (e.g. location, school size) and background (e.g. SAT).

  • Vision: Help high school students and parents make well-informed decisions in the college application process.
  • Mission: Create an interactive web app that is based on college information data to help applicants and their family better decide which colleges to apply for and to attend.
  • Success Criteria: Track new user engagement and interaction of the web app throughout the time.

Data

The raw data is from Kaggle. I used R to do some EDA and clean the raw data (code in develop/data_cleaning/data_cleaning.Rmd). Alternatively, you can download the cleaned data from my Google Drive.

Pivotal Tracker

Link to Pivotal Tracker

Software & Package requirements

Things you need to get it started:

  • conda: Either Anaconda or Miniconda is fine for this project.
  • git: You will most likely need version control.

Set up the app

Below is a brief tutorial to set up the app in a AWS EC2 or Linux. For other systems, the general steps should be the same, but small changes might be needed.

  1. Update. Install git and conda if you have not done so.

    sudo yum update

    sudo yum install git

    wget https://repo.continuum.io/archive/Anaconda3-5.1.0-Linux-x86_64.sh bash Anaconda3-5.1.0-Linux-x86_64.sh

  2. Clone this GitHub repository to local. Go into the directory, and use the collegeapp.yml file to create a conda environment with all required packages and dependecies.

    conda env create -f collegeapp.yml

    Then, activate the conda environment by entering source activate collegeapp.

  3. In the same directory as collegeapp.yml, create a file called config and paste the following information into the file to configure AWS RDS access.

    SECRET_KEY = 'development_key' SQLALCHEMY_DATABASE_URI = 'postgresql://collegeconnect:collegeahead@msiawebapp.cg96n7rbldvk.us-east-1.rds.amazonaws.com:5432/msiawebappdb' SQLALCHEMY_TRACK_MODIFICATIONS = True

  4. app/__init__.py should have included the following line of code:

    application.config.from_envvar('APP_SETTINGS', silent=True)

    which tells the application to look at the environmental variable APP_SETTINGS for the path to your config file. This means you simply need to set this environmental variable by entering:

    export APP_SETTINGS="path/to/where/your/config/file/is.config

  5. The database has been initialized, so you may skip this step. If this is not the case, please initialize a folder called data in develop and store the cleaned data (Google Drive) into this new folder. Then, you should enter python create_collegedb.py to initialize the database.

  6. Now enter python application.py. The app should be running on your EC2 public DNS + :5000/home. Have fun!

Logging

There are two sets of logging performed.

  1. application.log stores the logs of any user interaction with the EC2 application.

  2. createdb.log stores the logs of database initialization.

Unit Testing

We performed unit testing for develop/modeling/model.py file. The functions we tested are:

  • filter()
  • modeling()
  • major_pref_transformation()