Is there a way to get the historical text data for Central Bank Meetings' Minutes?
I am working on an NLP application and would like to know if a can get those texts from eikon or datastream in a smiliar way as you use ek.get_data or ds.get_data.
You can try getting these minutes from news. The following call retrieves news headlines using news topic code BRCO (Central bank of Columbia) and keyword 'minutes'. It returns a dataframe of news headlines related to Central bank of Columbia and containing the word 'minutes'.
hl_df = ek.get_news_headlines('BRCO AND MINUTES')
Then to get the story for the latest headline in the dataframe you can use
story = ek.get_news_story(hl_df['storyId'].iloc)
The story is returned as HTML. To parse the text out of it you can use BeatifulSoup library. The code snippet below removes all style and script tags from the HTML
from bs4 import BeautifulSoup soup = BeautifulSoup(story,"html.parser") for data in soup(['style', 'script']): data.decompose() story_text = ' '.join(soup.stripped_strings) print(story_text)
The result is
"After a thorough evaluation of economic activity, inflation and international financial conditions, Banco de la República's board of directors (BDBR) voted unanimously on June 28 to hold the benchmark interest rate at 1.75%...."