Update README.md

README.md

@@ -25,12 +25,12 @@ language:
 tags:
 - economics
 - finance
-- business
 size_categories:
 - 100K<n<1M
+pretty_name: central_bank_speeches_MLM
 ---
 # Dataset Card for "BIS_Speeches_97_23"
 
 This dataset is built from scrapped speeches on the Bank of International Settlements thanks to this repo : https://github.com/HanssonMagnus/scrape_bis. The dataset is made of 12k speeches from 1997 to 2023.
 Each pair is built with extracted sentences from speeches, if B is following A then the 'next_sentence_label' is 1 else it is 0.
-Negative pairs are built by choosing a sentence from another speech randomly.)
+Negative pairs are built by choosing a sentence from another speech randomly.