Dataset

stock-market-tweets-data

A community-compiled collection of stock-market-related tweets in English, stored as CSV, hundreds of thousands of posts (CC BY 4.0). Retail chatter feeds sentiment-signal research and market-tone studies.

Open on Hugging Face

Data preview

Subset

Loading preview...

Dataset card

stock-market-tweets-data is a publicly hosted mirror of a Twitter corpus collected between April 9 and July 16, 2020, using S&P 500 and Bloomberg-related hashtags and cashtags. It is distributed as a single CSV file under a CC BY 4.0 license, intended primarily for sentiment and text-classification research.

What this dataset contains

The corpus holds 923,673 training rows of English-language tweets pulled via the cashtags and hashtags listed by the publisher, including #SPX500, #SP500, #stocks, $SPX and individual tickers such as $AAPL, $MSFT, $AMZN, $GOOG, $JPM, $V, $PG, $MA, $BAC, $XOM, $DIS, $JNJ and others. The publisher describes the broader, unlabeled collection as 943,672 tweets and references a separate labeled subset of 1,300 tweets hosted elsewhere. No numeric fields beyond tweet identifiers are provided in the published schema.

Structure and fields

Rows are delivered in CSV form with three columns defined in the schema: id (int64), created_at (string timestamp) and text (string). The provided sample rows show id values starting at 1, created_at timestamps with the date masked, and text containing cashtags and financial vocabulary such as volume and Fibonacci levels. There are no explicit sentiment labels, ticker annotations or user identifiers in the published schema, so any per-tweet labeling must be added downstream.

How analysts use it

The publisher positions the dataset for academic and applied research on market-sentiment modeling, financial text classification and AI tooling for financial institutions, which is consistent with the 100K–1M size tag and the text-classification task category. Buy-side or systematic researchers could use it to prototype message-volume or keyword-based signals around S&P 500 constituents and macro tags such as #SPX500 and #stocks, but any production use would require additional filtering, retweet deduplication and bot screening not documented in the source.

Documentation & references 1