What does this research mean for the field?

A neural network model integrating deep semantic representation with contextual correlation analysis comprehensively outperforms mainstream baseline models in evaluating English-Chinese translation quality. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This study aims to develop a neural network model for effectively assessing the quality of English-Chinese text translations.

June 1, 2026International Journal of Information and Communication Technology0 citationsOpen Access

A neural network-based quality assessment model for English-to-Chinese text translation

胡胡铁江

Key Points

This study aims to develop a neural network model for effectively assessing the quality of English-Chinese text translations.
Developed a neural network model integrating deep semantic representation and contextual correlation analysis.
Evaluated on the WMT 2020 metrics shared task dataset for translation quality.
Utilized various quality metrics including Pearson's r and macro-F1 scoring.
Achieved a Pearson's r of 0.682 indicating strong correlation in quality assessment.
Obtained a macro-F1 score of 0.689 for multi-dimensional classification evaluation.
Achieved a normalised discounted cumulative gain@10 of 0.927, surpassing mainstream baseline models.

Abstract

Addressing the urgent need for cross-language text translation quality assessment, this paper proposes a neural network-based model for evaluating English-Chinese translation quality.Current widely adopted automated evaluation methods exhibit significant limitations in handling specialised terminology and nuanced semantics, particularly when addressing culture-specific concepts.The neural network model constructed in this study integrates deep semantic representation with contextual correlation analysis, achieving remarkable results on the Chinese-English test set of the public WMT 2020 metrics shared task dataset.It achieved a core correlation metric (Pearson's r) of 0.682, along with a multi-dimensional classification evaluation (macro-F1) of 0.689 and a ranking quality metric (normalised discounted cumulative gain @10) of 0.927, comprehensively outperforming mainstream baseline models.This model provides a reliable technical tool for cross-language text quality control.

问 AI

Bookmark

View Full Paper