I'm learning sentiment analysis. I need financial tweets and blogs dataset for supervised learning.
Right now I'm trying lexicon based sentiment analysis on a small dataset of financial tweets from stocktwits.

Can someone guide me where to find other sets?

2 Answers 2


467 million Twitter posts from 20 million users covering a 7 month period from June 1 2009 to December 31 2009. We estimate this is about 20-30% of all public tweets published on Twitter during the particular time frame.

For each public tweet the following information is available:

Author Time Content

Please refer this link: https://snap.stanford.edu/data/other.html

  • Good timing. Twitter introduced the $AAPL notation to indicate a stock in February, 2009 wired.com/2009/02/on-twitter-is-t (this should help you filter out tweets mentioning stocks)
    – philshem
    Mar 18, 2018 at 13:56

I would be tempted to use Google Finance or Yahoo Finance to get a list of stock symbols. i.e. VOD.L

Then you could use the twitter API to extract search results for each one, again i.e. $VOD.L https://twitter.com/search?q=%24VOD.L

There is a list of news APIs on Programmable Web that could perhaps be used to extract news items for each symbol: http://www.programmableweb.com/category/News%20Services/apis?category=20250

From those you should be able to extract past tweets/news articles as well as historical pricing data and then generate your own training sets.

  • I originally thought of this solution as a last resort, if i don't get the tagged dataset. Thanks for the APIs search engine, it's a good resource.
    – pauli
    Apr 6, 2016 at 2:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.