Dataset Card for STAN Large

Dataset Summary

The description below was taken from the paper "Multi-task Pairwise Neural Ranking for Hashtag Segmentation" by Maddela et al..

"STAN large, our new expert curated dataset, which includes all 12,594 unique English hashtags and their associated tweets from the same Stanford dataset.

STAN small is the most commonly used dataset in previous work. However, after reexamination, we found annotation errors in 6.8% of the hashtags in this dataset, which is significant given that the error rate of the state-of-the art models is only around 10%. Most of the errors were related to named entities. For example, #lionhead, which refers to the “Lionhead” video game company, was labeled as “lion head”.

We therefore constructed the STAN large dataset of 12,594 hashtags with additional quality control for human annotations."

Languages

English

Dataset Structure

Data Instances

{
    "index": 15,
    "hashtag": "PokemonPlatinum",
    "segmentation": "Pokemon Platinum",
    "alternatives": {
        "segmentation": [
            "Pokemon platinum"
        ]
    }
}

Data Fields

• index : a numerical index.
• hashtag : the original hashtag.
• segmentation : the gold segmentation for the hashtag.
• alternatives : other segmentations that are also accepted as a gold segmentation.

Although segmentation has exactly the same characters as hashtag except for the spaces, the segmentations inside alternatives may have characters corrected to uppercase.

Dataset Creation

• All hashtag segmentation and identifier splitting datasets on this profile have the same basic fields: hashtag and segmentation or identifier and segmentation .

• The only difference between hashtag and segmentation or between identifier and segmentation are the whitespace characters. Spell checking, expanding abbreviations or correcting characters to uppercase go into other fields.

• There is always whitespace between an alphanumeric character and a sequence of any special characters ( such as _ , : , ~ ).

• If there are any annotations for named entity recognition and other token classification tasks, they are given in a spans field.

Additional Information

Citation Information

@inproceedings{maddela-etal-2019-multi,
    title = "Multi-task Pairwise Neural Ranking for Hashtag Segmentation",
    author = "Maddela, Mounica  and
      Xu, Wei  and
      Preo{\c{t}}iuc-Pietro, Daniel",
    booktitle = "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics",
    month = jul,
    year = "2019",
    address = "Florence, Italy",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/P19-1242",
    doi = "10.18653/v1/P19-1242",
    pages = "2538--2549",
    abstract = "Hashtags are often employed on social media and beyond to add metadata to a textual utterance with the goal of increasing discoverability, aiding search, or providing additional semantics. However, the semantic content of hashtags is not straightforward to infer as these represent ad-hoc conventions which frequently include multiple words joined together and can include abbreviations and unorthodox spellings. We build a dataset of 12,594 hashtags split into individual segments and propose a set of approaches for hashtag segmentation by framing it as a pairwise ranking problem between candidate segmentations. Our novel neural approaches demonstrate 24.6{\%} error reduction in hashtag segmentation accuracy compared to the current state-of-the-art method. Finally, we demonstrate that a deeper understanding of hashtag semantics obtained through segmentation is useful for downstream applications such as sentiment analysis, for which we achieved a 2.6{\%} increase in average recall on the SemEval 2017 sentiment analysis dataset.",
}

Contributions

This dataset was added by @ruanchaves while developing the hashformers library.