CLAug 22, 2024

CLEANANERCorp: Identifying and Correcting Incorrect Labels in the ANERcorp Dataset

arXiv:2408.12362v280 citationsh-index: 13
Originality Synthesis-oriented
AI Analysis

This addresses label errors that can harm model training and evaluation for researchers using Arabic NER datasets, but it is incremental as it focuses on correcting an existing dataset.

The study identified and corrected annotation errors, missing labels, and inconsistencies in the ANERcorp dataset for Arabic Named Entity Recognition, resulting in a cleaner version called CLEANANERCorp to provide a more accurate benchmark.

Label errors are a common issue in machine learning datasets, particularly for tasks such as Named Entity Recognition. Such label errors might hurt model training, affect evaluation results, and lead to an inaccurate assessment of model performance. In this study, we dived deep into one of the widely adopted Arabic NER benchmark datasets (ANERcorp) and found a significant number of annotation errors, missing labels, and inconsistencies. Therefore, in this study, we conducted empirical research to understand these errors, correct them and propose a cleaner version of the dataset named CLEANANERCorp. CLEANANERCorp will serve the research community as a more accurate and consistent benchmark.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes