BERT for stock market sentiment analysis

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Authors

  • Matheus Gomes Sousa
  • Kenzo Sakiyama
  • Lucas De Souza Rodrigues
  • Pedro Henrique Moraes
  • Eraldo Fernandes
  • Edson Takashi Matsubara

When breaking news occurs, stock quotes can change abruptly in a matter of seconds. The human analysis of breaking news can take several minutes, and investors in the financial markets need to make quick decisions. Such challenging scenarios require faster ways to support investors. In this work, we propose the use of bidirectional encoder representations from transformers BERT to perform sentiment analysis of news articles and provide relevant information for decision making in the stock market. This model is pre-trained on a large amount of general-domain documents by means of a self-learning task. To fine-tune this powerful model on sentiment analysis for the stock market, we manually labeled stock news articles as positive, neutral or negative. This dataset is freely available and amounts to 582 documents from several financial news sources. We fine-tune a BERT model on this dataset and achieve 72.5% of F-score. Then, we perform some experiments highlighting how the output of the obtained model can provide valuable information to predict the subsequent movements of the Dow Jones Industrial (DJI) Index.

Original languageEnglish
Title of host publicationIEEE 31st International Conference on Tools with Artificial Intelligence : ICTAI 2019 : proceedings : 4-6 November 2019, Portland, Oregon
Number of pages5
Place of PublicationPiscataway
PublisherIEEE - Institute of Electrical and Electronics Engineers Inc.
Publication date11.2019
Pages1589-1593
ISBN (print)978-1-7281-3799-5
ISBN (electronic)978-1-7281-3798-8
DOIs
Publication statusPublished - 11.2019
Externally publishedYes
EventIEEE International Conference on Tools with Artificial Intelligence - Portland, United States
Duration: 04.11.201906.11.2019
Conference number: 13
https://ictai.computer.org/

DOI

Recently viewed

Researchers

  1. Joshua Simon

Publications

  1. Implementing Corporate Sustainability
  2. Learning Soccer in Elementary School: Using Teaching Games for Understanding and Digital Media
  3. Entangled Identities
  4. Die Macht der Funktionen
  5. Improving the cost-effectiveness of a healthcare system for depressive disorders by implementing telemedicine
  6. New Media, Old Media: A History and Theory Reader
  7. Influence of clouds on the photochemical degradation in aqueous phase - II
  8. Web-based occupational stress prevention in German micro- and small-sized enterprises – process evaluation results of an implementation study
  9. A meta-analytic reliability generalization of the Physical Self-Description Questionnaire (PSDQ)
  10. A target costing approach to developing an online distribution channel
  11. I'm lonely, can't you tell?
  12. Variational pragmatics
  13. Smelling like resin
  14. Friede den Völkern
  15. Cycling at varying load
  16. Konfiguration der PPS
  17. Measuring Method Effects
  18. Coauthoring collaborative strategy when voices are many and authority is ambiguous
  19. Friction riveting of 3D printed polyamide 6 with AA 6056-T6
  20. Site use of grazing cattle and sheep in a large-scale pasture landscape
  21. Quality management in a top tier accounting firm: Towards a socio-cognitive model
  22. Enterprise Architecture as a Tool for Managing Corporate Social Responsibility
  23. An individual differences perspective on pragmatic abilities in the preschool years
  24. Indentity 2.0
  25. Über das Essen
  26. Trust-Based Cooperation Relationships between SMEs
  27. Exploring the “works with nature” pillar of food sovereignty
  28. Identity-b(i)ased intervention of third-parties
  29. Im Schatten des Vielfaltsdiskurses
  30. Rechtschreiblernen in der Sekundarstufe I