BomJi at SemEval-2018 Task 10: Combining Vector-, Pattern- and Graph-based Information to Identify Discriminative Attributes

TL;DR

This paper presents BomJi, a supervised system that combines multiple feature types to identify discriminative attributes in word pairs, achieving high performance in a SemEval task.

Contribution

The system integrates vector-, pattern-, and graph-based features with an XGB classifier, demonstrating effective feature engineering for discriminative attribute detection.

Findings

Achieved an F1 score of 0.73 on the task

Ranked 2nd out of 26 systems

Effective combination of diverse feature types

Abstract

This paper describes BomJi, a supervised system for capturing discriminative attributes in word pairs (e.g. yellow as discriminative for banana over watermelon). The system relies on an XGB classifier trained on carefully engineered graph-, pattern- and word embedding based features. It participated in the SemEval- 2018 Task 10 on Capturing Discriminative Attributes, achieving an F1 score of 0:73 and ranking 2nd out of 26 participant systems.