PyFin-sentiment: Towards a machine-learning-based model for deriving sentiment from financial tweets

Research output: Journal contributionsJournal articlesResearchpeer-review

Authors

Responding to the poor performance of generic automated sentiment analysis solutions on domain-specific texts, we collect a dataset of 10,000 tweets discussing the topics of finance and investing. We manually assign each tweet its market sentiment, i.e., the investor's anticipation of a stock's future return. Using this data, we show that all existing sentiment models trained on adjacent domains struggle with accurate market sentiment analysis due to the task's specialized vocabulary. Consequently, we design, train, and deploy our own sentiment model. It outperforms all previous models (VADER, NTUSD-Fin, FinBERT, TwitterRoBERTa) when evaluated on Twitter posts. On posts from a different platform, our model performs on par with BERT-based large language models. We achieve this result at a fraction of the training and inference costs due to the model's simple design. We publish the artifact as a python library to facilitate its use by future researchers and practitioners.

Original languageEnglish
Article number100171
JournalInternational Journal of Information Management Data Insights
Volume3
Issue number1
Number of pages10
DOIs
Publication statusPublished - 01.04.2023
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 2023 The Author(s)

Recently viewed

Researchers

  1. Lea Wollschläger

Publications

  1. Turbulente Ränder
  2. Added value of convection-permitting simulations for understanding future urban humidity extremes
  3. Red List of marine macroalgae of the Wadden Sea
  4. Experimentation for Sustainable Innovation
  5. Propagating Maximum Capacities for Recommendation
  6. Differentiated Instruction Around the World - A Global Inclusive Insight
  7. Calculation of Physicochemical Properties for Short- and Medium-Chain Chlorinated Paraffins
  8. Log in and breathe out: internet-based recovery training for sleepless employees with work-related strain
  9. Asset Backed Securities
  10. New concepts of extrusion dies to reduce the anisotropy of extruded profiles by means of additive manufacturing
  11. Corporate social responsibility performance, reporting and generalized methods of moments (GMM)
  12. Faszination Programmierung
  13. Guest editorial
  14. Online hands-on trainings (real worlds in virtual environments)
  15. Traits of butterfly communities change from specialist to generalist characteristics with increasing land-use intensity
  16. Host plant availability potentially limits butterfly distributions under cold environmental conditions
  17. Land use change and the future of biodiversity
  18. Die Welteislehre
  19. Evidence-Based Management
  20. Group membership does not modulate automatic imitation
  21. Editorial zum Themenschwerpunkt
  22. Bundesrat
  23. Bunker schreiben
  24. Labour Market Participation of Older Workers
  25. Benno Reifenberg (1892-1970)
  26. Foreign and Domestic Takeovers in Germany: First Comparative Evidence on the Post-acquisition Target Performance using new Data
  27. How passion in entrepreneurship develops over time
  28. Investigation of interaction between forming processes and rotor geometries of screw machines