Summary of Wikiner-fr-gold: a Gold-standard Ner Corpus, by Danrun Cao (irisa et al.
WikiNER-fr-gold: A Gold-Standard NER Corpus
by Danrun Cao, Nicolas Béchet, Pierre-François Marteau
First submitted to arxiv on: 29 Oct 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Databases (cs.DB)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed paper addresses the quality of the WikiNER corpus, a multilingual Named Entity Recognition (NER) corpus, by providing a consolidated version of it. The original annotation process was semi-supervised, resulting in a “silver-standard” corpus. To improve the accuracy of the French portion of WikiNER, the authors propose WikiNER-fr-gold, which represents 20% of the original French sub-corpus (26,818 sentences with 700k tokens). The paper outlines an annotation guideline and presents an analysis of errors and inconsistencies observed in the original WikiNER-fr corpus. Future work directions are also discussed. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper talks about a big dataset for recognizing named entities like people, places, and organizations across many languages. This dataset is important because it can help computers understand text better. The problem with this dataset is that it wasn’t checked carefully before being released, which means there might be mistakes in it. To fix this, the authors created a new version of the French part of the dataset that’s more accurate. They looked at how the original data was labeled and fixed some errors. This makes the dataset better for people who want to use it to teach computers about named entities. |
Keywords
» Artificial intelligence » Named entity recognition » Ner » Semi supervised
