BomJi at SemEval-2018 Task 10: Combining Vector-, Pattern- and Graph-based Information to Identify Discriminative Attributes
Enrico Santus, Chris Biemann, Emmanuele Chersoni
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.
Results
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Relation Extraction | SemEval 2018 Task 10 | F1-Score | 0.73 | Gradient boosting with co-occurrence count features and JoBimText features |
Related Papers
DocIE@XLLM25: In-Context Learning for Information Extraction using Fully Synthetic Demonstrations2025-07-08Multiple Streams of Relation Extraction: Enriching and Recalling in Transformers2025-06-25Chaining Event Spans for Temporal Relation Grounding2025-06-17Summarization for Generative Relation Extraction in the Microbiome Domain2025-06-10Conservative Bias in Large Language Models: Measuring Relation Predictions2025-06-09Comparative Analysis of AI Agent Architectures for Entity Relationship Classification2025-06-03CREFT: Sequential Multi-Agent LLM for Character Relation Extraction2025-05-30Generating Diverse Training Samples for Relation Extraction with Large Language Models2025-05-29