Automatic Text Scoring Using Neural Networks

TL;DR

This paper presents a neural network-based automated text scoring system that learns word contributions and uses LSTM to represent text meaning, achieving high performance and interpretability.

Contribution

It introduces a fully automated neural framework for text scoring that learns word importance and visualizes discriminative text regions.

Findings

Achieves excellent scoring accuracy compared to similar methods.

Provides a novel interpretability method for neural text scoring models.

Demonstrates the effectiveness of LSTM-based representations in ATS.

Abstract

Automated Text Scoring (ATS) provides a cost-effective and consistent alternative to human marking. However, in order to achieve good performance, the predictive features of the system need to be manually engineered by human experts. We introduce a model that forms word representations by learning the extent to which specific words contribute to the text's score. Using Long-Short Term Memory networks to represent the meaning of texts, we demonstrate that a fully automated framework is able to achieve excellent results over similar approaches. In an attempt to make our results more interpretable, and inspired by recent advances in visualizing neural networks, we introduce a novel method for identifying the regions of the text that the model has found more discriminative.