Skip to content
No description, website, or topics provided.
Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
docs
src
.gitignore
.gitmodules
LICENSE
README.md

README.md


Automated-Difficulty-prediction-for-Exam-Questions

Motivation


  • To address the challenges faced in ensuring authenticity and fairness for Test-Takers to who attempt Online Exams.
  • The challenge is to ensure each question set for test-taker contains questions of similar difficulty levels.
  • Come-up with a web-application which will ensure that the questions are tagged with correct difficulty level for next iteration of Exam, using machine learning analysis on the performance of test-takers for previous exam.

Ojectives achieved


  • Identification of the features in the Exam Question Data
  • Development of Machine Learning Algorithm for predicting the difficulty levels or tags of the Question Data
  • Provide User Interface for Question Creators with following features:
    • Provision for performing Machine Learning analysis on Selected Exam Data
    • Creation of Analysis report on difficulty tagging for each year‘s Exam data
    • Statistics for change in the tags, that is, deficits and excess, overall
    • Adding new Questions
    • Editing Previous Questions
    • Provided Question Revision history and provision to mark question active or inactive
    • Provided Audit trail on Questions
    • Question Sets creation with the option of shuffling and pagination
    • User Management for Admin, Question Editor and Moderator

Implementation


Workflow

Usecase

Documentation


Report and documentation can be found on this Documentation

Folder Tree


  • docs contains documentation and paper
  • src contains codes
    • 611-Proj-Ques-Eval-Frontend: Contains the Website
    • Python-Web-Server: Provide REST APIs solution for running Machine Learning Algorithm

Contributors


Instructor


Prof. Alan Hunt, CSE Dept. University at Buffalo - SUNY

Tech-stack

  • Machine Learning
    • Python 3.6
    • Tornado for Rest APIs
    • SQLAlchemy
    • Matplotlib for plotting the clusters
  • Web Development
    • Laravel 5.4 with Symphony and Eloquent ORM
    • PHP 7
    • MySQL
    • Apache Server for deploying the Website
    • Bootstrap 3.6
    • Google Charts

Acknowledgement


  • e-Yantra, IIT-Bombay

License


This project is open-sourced under MIT License

You can’t perform that action at this time.