Open Dataset
Data Structure ?
243.44M
Data Structure ?
*The above analysis is the result extracted and analyzed by the system, and the specific actual data shall prevail.
README.md
***コンテキスト***
このデータセットには、インターネットのコメントサイトRedditから収集された130万件の皮肉なコメントが含まれています。このデータセットは、Redditから(私ではなく:)(皮肉) タグが付いたコメントをスクレイピングすることで生成されました。このタグは、Redditのユーザーが自分のコメントが冗談であり、真剣に受け取られることを意図していないことを示すためによく使われ、一般的に皮肉なコメント内容の信頼できる指標となります。
***内容***
データには、均衡型と不均衡型(すなわち実際の分布)のバージョンがあります。(実際の比率は約1:100です。)コーパスには130万件の皮肉な発言が含まれており、それらが返信した内容や、同じソースからの多くの非皮肉なコメントも含まれています。
ラベル付けされたコメントは、train - balanced - sarcasm.csvファイルにあります。
***謝辞***
このデータは、Mikhail Khodak、Nikunj Saunshi、Kiran Vodrahalliが彼らの論文「A Large Self - Annotated Corpus for Sarcasm」のために収集したものです。データはここにホストされています。
***引用:***
@unpublished{SARC,
authors={Mikhail Khodak and Nikunj Saunshi and Kiran Vodrahalli},
title={A Large Self - Annotated Corpus for Sarcasm},
url={https://arxiv.org/abs/1704.05579},
year=2017
}
元のデータセット内のファイルの注釈:readme.txt。
***着想***
皮肉の予測と関連する自然言語処理の特徴(例:主観的決定要因、人種差別、条件文、感情的な単語、「インターネットスラング」や特定のフレーズ)。
皮肉と感情の比較
大文字、イタリック、または伸長された単語などの異常な言語的特徴。例:「Yeahhh, I'm sure THAT is the right answer」(そうだね、それが正解だと確信してるわ)。
人々が皮肉を持って反応しやすいトピック
×
The dataset is currently being organized and other channels have been prepared for you. Please use them
The dataset is currently being organized and other channels have been prepared for you. Please use them
Note: Some data is currently being processed and cannot be directly downloaded. We kindly ask for your understanding and support.
No content available at the moment
No content available at the moment
- Share your thoughts
Go share your ideas~~
ALL
Welcome to exchange and share
Your sharing can help others better utilize data.
Data usage instructions: h1>
I. Data Source and Display Explanation:
- 1. The data originates from internet data collection or provided by service providers, and this platform offers users the ability to view and browse datasets.
- 2. This platform serves only as a basic information display for datasets, including but not limited to image, text, video, and audio file types.
- 3. Basic dataset information comes from the original data source or the information provided by the data provider. If there are discrepancies in the dataset description, please refer to the original data source or service provider's address.
II. Ownership Explanation:
- 1. All datasets on this site are copyrighted by their original publishers or data providers.
III. Data Reposting Explanation:
- 1. If you need to repost data from this site, please retain the original data source URL and related copyright notices.
IV. Infringement and Handling Explanation:
- 1. If any data on this site involves infringement, please contact us promptly, and we will arrange for the data to be taken offline.
- 1. The data originates from internet data collection or provided by service providers, and this platform offers users the ability to view and browse datasets.
- 2. This platform serves only as a basic information display for datasets, including but not limited to image, text, video, and audio file types.
- 3. Basic dataset information comes from the original data source or the information provided by the data provider. If there are discrepancies in the dataset description, please refer to the original data source or service provider's address.
- 1. All datasets on this site are copyrighted by their original publishers or data providers.
- 1. If you need to repost data from this site, please retain the original data source URL and related copyright notices.
- 1. If any data on this site involves infringement, please contact us promptly, and we will arrange for the data to be taken offline.