NLP-CIC @ PRELEARN: Mastering prerequisites relations, from handcrafted features to embeddings

TL;DR

This paper describes a system for classifying prerequisite relations between concepts, achieving top performance using handcrafted features and embeddings across different domains.

Contribution

It introduces a high-performing approach combining handcrafted features and embeddings for prerequisite relation classification, winning EVALITA 2020.

Findings

Ranked first in both in-domain and cross-domain scenarios.

Achieved average F1 scores of 0.887 and 0.690.

Code is publicly available.

Abstract

We present our systems and findings for the prerequisite relation learning task (PRELEARN) at EVALITA 2020. The task aims to classify whether a pair of concepts hold a prerequisite relation or not. We model the problem using handcrafted features and embedding representations for in-domain and cross-domain scenarios. Our submissions ranked first place in both scenarios with average F1 score of 0.887 and 0.690 respectively across domains on the test sets. We made our code is freely available.