arXiv Open Access 2018

Human vs Automatic Metrics: on the Importance of Correlation Design

Anastasia Shimorina

Lihat Sumber

Abstrak

This paper discusses two existing approaches to the correlation analysis between automatic evaluation metrics and human scores in the area of natural language generation. Our experiments show that depending on the usage of a system- or sentence-level correlation analysis, correlation results between automatic scores and human judgments are inconsistent.

Topik & Kata Kunci

cs.CL

Penulis (1)

Anastasia Shimorina

Format Sitasi

APA MLA BibTeX

Shimorina, A. (2018). Human vs Automatic Metrics: on the Importance of Correlation Design. https://arxiv.org/abs/1805.11474

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2018
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓