Reference and Document Aware Semantic Evaluation Methods for Korean Language Summarization

TL;DR

This paper introduces RDASS, a semantic evaluation metric tailored for Korean summarization, addressing limitations of ROUGE by better reflecting semantic content and improving correlation with human judgment.

Contribution

The paper proposes RDASS, a novel semantic evaluation metric for Korean summarization, and a method to enhance its alignment with human assessments.

Findings

RDASS shows higher correlation with human judgment than ROUGE.

The proposed method improves the semantic evaluation accuracy for Korean summaries.

Evaluation results demonstrate the effectiveness of RDASS in Korean language summarization.

Abstract

Text summarization refers to the process that generates a shorter form of text from the source document preserving salient information. Many existing works for text summarization are generally evaluated by using recall-oriented understudy for gisting evaluation (ROUGE) scores. However, as ROUGE scores are computed based on n-gram overlap, they do not reflect semantic meaning correspondences between generated and reference summaries. Because Korean is an agglutinative language that combines various morphemes into a word that express several meanings, ROUGE is not suitable for Korean summarization. In this paper, we propose evaluation metrics that reflect semantic meanings of a reference summary and the original document, Reference and Document Aware Semantic Score (RDASS). We then propose a method for improving the correlation of the metrics with human judgment. Evaluation results show that the correlation with human judgment is significantly higher for our evaluation metrics than for ROUGE scores.