Repository for the ACL 2022 paper:
Incorporating Stock Market Signals for Twitter Stance Detection

Link to the paper: Incorporating Stock Market Signals for Twitter Stance Detection

Please use the following citation:

@inproceedings{conforti2022stocks,
          title={Incorporating Stock Market Signals for Twitter Stance Detection},
          author={Conforti, Costanza and Berndt, Jakob and Pilehvar, Mohammad Taher and Giannitsarou, Chryssi and Toxvaerd, Flavio and Collier, Nigel} 
          booktitle={Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (ACL2022)},
          year={2022}
        }

Dataset

This data extend the Will-They-Won't-They (WT-WT) dataset. WT-WT is a large dataset of English tweets targeted at stance detection for the rumor verification task. The dataset is constructed based on tweets that discuss five recent merger and acquisition (M&A) operations of US companies, mainly from the healthcare sector.

For the four healthcare mergers, we obtain historical prices in 30-min intervals for the involved stocks. Each entry in the data has the following fields: DateTime, Open, High, Low, Close, Volume. DateTime is in US Eastern Time, in the format YY-MM-DD h:m:s. Only minutes with trading volume are included: times with zero volume, such as during weekends or holidays, are omitted. Prices are adjusted for dividends and splits.

Considered M&A operations

Operation	Buyer	Target	Industry
CVS_AET	CVS Health	Aetna	Healthcare
CI_ESRX	Cigna	Express Scripts	Healthcare
ANTM_CI	Anthem	Cigna	Healthcare
AET_HUM	Aetna	Humana	Healthcare

Contacts

• {cc918, jb2088, mp792, cg349, fmot2, nhc30} @cam.ac.uk
• Cambridge Language Technology Lab
• Cambridge Faculty of Economics