Discovery News: A Generic Framework for Financial News Recommendation


  • Chong Wang S&P Global Ratings
  • Lisa Kim S&P Global Ratings
  • Grace Bang S&P Global Ratings
  • Himani Singh S&P Global Ratings
  • Russell Kociuba S&P Global Ratings
  • Steven Pomerville S&P Global Ratings
  • Xiaomo Liu S&P Global Ratings



In the financial services industry, it is crucial for analysts to constantly monitor and stay informed on the latest developments of their portfolio of companies. This ensures that analysts are up-to-date in their analysis and provide highly credible and timely insights. Currently, analysts receive news alerts through manually created news alert subscriptions that are often noisy and difficult to manage. The manual review process is time-consuming and error-prone. We demonstrate Discovery News, a framework for an automated news recommender system for financial analysis at S&P's Global Ratings. This system includes the automated ingestion, relevancy, clustering, and ranking of news. The proposed framework is adaptable to any form of input news data and can seamlessly integrate with other data used for analysis like financial data.




How to Cite

Wang, C., Kim, L., Bang, G., Singh, H., Kociuba, R., Pomerville, S., & Liu, X. (2020). Discovery News: A Generic Framework for Financial News Recommendation. Proceedings of the AAAI Conference on Artificial Intelligence, 34(08), 13390-13395.



IAAI Technical Track: Emerging Papers