Machine Comprehension Based on Learning to Rank

TL;DR

This paper explores a feature-engineered approach using semantics and learning to rank for machine comprehension, achieving efficient performance with less training data on a large news dataset.

Contribution

It introduces a learning to rank based system that improves machine comprehension efficiency and reduces training data requirements compared to deep learning methods.

Findings

L2R system achieves competitive performance

Semantic features enhance comprehension accuracy

Less training data needed for effective results

Abstract

Machine comprehension plays an essential role in NLP and has been widely explored with dataset like MCTest. However, this dataset is too simple and too small for learning true reasoning abilities. \cite{hermann2015teaching} therefore release a large scale news article dataset and propose a deep LSTM reader system for machine comprehension. However, the training process is expensive. We therefore try feature-engineered approach with semantics on the new dataset to see how traditional machine learning technique and semantics can help with machine comprehension. Meanwhile, our proposed L2R reader system achieves good performance with efficiency and less training data.