reuters-dataset

🗞️ A tool for downloading and parsing Reuters-21578. These are a collection of documents that appeared on Reuters newswire back in 1987.

🔥 Features

• Asynchronously caches the full dataset to your temporary directory.
  • This reduces your project size.
• Prettifies the results.
  • Uses proper JSON naming conventions and common-sense values.

🚀 Getting Started

Using npm:

npm install --save reuters-dataset

Using yarn:

yarn add reuters-dataset

✍️ Usage

import getReutersDataset from 'reuters-dataset';

(
  async () => {
    const { exchanges, orgs, people, places, topics, articles } = await getReutersDataset();
  }
)();

📌 Example

{
  "$": {
    "topics": true,
    "lewissplit": "TRAIN",
    "cgisplit": "TRAINING-SET",
    "oldid": "5544",
    "newid": "1"
  },
  "topics": ["cocoa"],
  "places": ["el-salvador", "usa", "uruguay"],
  "people": [],
  "orgs": [],
  "exchanges": [],
  "companies": [],
  "text": {
    "title": "BAHIA COCOA REVIEW",
    "dateline": "SALVADOR, Feb 26 -",
    "body": "Showers continued throughout [...]"
  },
  "date": "1987-02-26T15:01:01.790Z"
}

✌️ License

MIT