These are the scripts used for the getting the dataset shown here: https://www.kaggle.com/miguelaenlle/reuters-articles-for-3500-stocks-since-2017

All the prerequisite files are in requirements.txt

1. Download this repository
2. Install the requirements via $ pip install -r requirements.txt

reuters_scraper.py is for getting scrape data for individual stocks. reuters_full_database_scraper.py is for scraping the entire Reuters database.

To use reuters_scraper.py,

• Put your script in this directory
• import reuters_scraper

reuters_scraper has two functions:

• get_data_for_stock
• get_data_for_stock_with_lookback

get_data_for_stock takes 2 arguments: stock and verbose If verbose is False, get_data_for_stock will not print anything. get_data_for_stock gets the entire reuters article history for the stock. Use this for backtesting.

get_data_for_stock_with_lookback also takes 2 arguments: stock and lookback It behaves similarly to get_data_for_stock, but lookback controls the amount of articles you want. Lookback is the amount of days backward the earliest publish date for an article should be, e.g. 7 day lookback = Get articles for the past 7 days

To use reuters_full_database_scraper.py, simply call the script via python reuters_full_database_scraper.py and follow the prompts shown on the command line.