Course Recommendation System

Detailed report available here!

A system that will help in a personalized recommendation of courses for an upcoming semester based on the performance of previous semesters.

For running individual models

1. User Based: $ python user.py
2. Item Based: $ python item.py
3. Warp MF: $ python hmf_warp_log.py
4. Logistic MF: $ python hmf_warp_log.py
5. Auto encoders: $ python auto_enc.py

It will return top 5 recommended courses and their grades. Also it will return top 5 courses that are compared with ground truth.

For running the web app locally on your system do

1. $ node main.js
2. Go to the browser and browse http://localhost:3000/
3. Enter the student ID and Semester for which you want top 5 courses alongwith their the grades.

matrix_creation.py is used to import the data into different models. Used as an import file.

Packages need to be installed: LightFM, sklearn, theano, pandas, numpy, scipy, npm (express, body-parser and ejs modules for nodejs).