'''

• Title: "Facebook Users Profiling "

• Category: Machine Learning

• Authors: Ryan Mokarian, Maysam Mokarian, Maziar M-Shahi

Description:
Facebook users’ various characteristics (age group, gender and personality traits) are predicted. Users’ posts (LIWC, NRC), page likes and images (Oxford) are utilized multimodal datasets. For Gender prediction, Random forest used. For Age group prediction, Logistic Regression used and for Personality traits prediction, Linear Regression and Ridge CV used. 

'''

• Overview
• Set up your environment
• Architecture

This repository contains the source code for the ift6758 course.

Follow the below steps to set up your virtual environment:

1. Make sure that you have installed on your local machine:

   pip install virtualenv

2. Create a virtual environment:

   virtualenv venv

3. Activate the virtual environment:

   • On Windows:

   venv/Scripts/activate

   • On Mac/Linux

   source venv/bin/activate

4. Install the project requirements on the virtual environment:

   pip install -r requirements.txt

Assuming you have the Train/ test data as below in the root of the project:

• data/Public_Test

  • Image
    • oxford.csv
  • Profile
    • Profile.csv
  • Relation
    • Relation.csv
  • Text
    • liwc.csv
    • nrc.csv

• data/Train

  • Image
    • oxford.csv
  • Profile
    • Profile.csv
  • Relation
    • Relation.csv
  • Text
    • liwc.csv
    • nrc.csv

Follow below steps to generate the models and predict the test data

• Navigate to source directory: cd src

• generate model: python model_generator.py

• Now you have models generated, you can use test data to predict and generate xml files (under data/results)

• predict: python resultgenerator.py

Below is the architecture of the system.