datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
stocktwits-cryptoFor academic reference, cite the following paper: https://ieeexplore.ieee.org/document/10223689
Dataset StockTwits-crypto contains all cryptocurrency-related posts from the StockTwits website, from 1st of November 2021 to the 15th of June 2022.
The data has been cleaned and preprocessed, we removed:
cashtags, hashtags, usernames,
URLs, crypto wallets,
Chinese, Korean and Japanese characters,
(most) UTF-8 encoding issues
removed all posts shorter than 4 words
removed all duplicate posts… See the full description on the dataset page: https://huggingface.co/datasets/ElKulako/stocktwits-crypto.stocktwits-emojiFor academic reference, cite the following paper: https://ieeexplore.ieee.org/document/10223689
This data set contains StockTwits posts from 01.11.2021 to 30.06.2022 for Bitcoin (BTC.X), Ethereum (ETH.X) and Shiba Inu (SHIB.X).
The full set contains 124,503 posts, including 72,247 bullish, 38,249 neutral and 14,007 bearish posts.
The training set ranges from 01.11.2021 to 30.04.2022, consists of 91,758 observations, including 57,932 bullish, 26,516 neutral, and 7310 bearish posts.
The… See the full description on the dataset page: https://huggingface.co/datasets/ElKulako/stocktwits-emoji.stocktwits_volatilitystocktwits_volatilitystocktwits_volatility_rawstocktwits_volatility_raw