Automatic Short Answer Grading in an Online Quiz System
Abstract
This thesis researches the problem of Automatic Short Answer Grading (ASAG) in order to build an automatic grading module for short-text responses. This module is developed for an online Learning Management System (LMS) called Stembord. A Multi-layer Perceptron model is the best performing model based on the features obtained during research. This research also informs the design and implementation of the short-answer grading module which is consequently integrated into an API service and called by the Stembord application.