NLP for Financial Markets

Goals

The Natural Language Processing (NLP) for Financial Markets team works on the development and application of NLP models for Financial Markets Including but not limited to Equity Market, Bond Market, and Cryptocurrency Market. Modern techniques help to better understand and predict financial market movements for investment in various financial markets through state-of-the-art NLP like FLANG-BERT, ChatGPT, etc.

Issues Involved or Addressed

With the application of NLP in financial markets comes the natural challenge of the need to collect and clean large datasets before applying any model. Many projects under the team require substantial data scraping, data cleaning, data labeling, and data wrangling. We try to optimize all data-related tasks using the latest ideas for example data labeling tasks can be optimized by employing weak-supervision techniques. Once we convert unstructured data to a more structured machine-readable form, we apply the latest ML/NLP models to answer essential questions about economics, and various financial markets. In the past students in our team worked on a wide range of problems including, “Using ChatGPT-based Sentiment Analysis on Consumer Reviews to Predict Price Changes of Products on Amazon”, “Investor Insight: Exploring Companies’ Annual SEC Reports through Knowledge Graphs”, “How Automobile Companies Change Narrative Based on EV Policy Change”, “Understanding Effect of Inflation on Real Estate Industry through NLP”, etc.

Partners/Sponsors

Financial Services and Innovation Lab, Master of Science in Quantitative and Computational Finance (MS-QCF), Partnership for an Advanced Computing Environment (PACE), Cloud Hub for IDEaS