Financial Modeling based on Sentiment Analysis and Natural Language Processing

Financial variables modelling plays an essential role in computational finance and risk management. Recent research has shown that public sentiment and other information expressed in the natural text such as news articles are important factors correlated with financial variables. The main purpose of this project is to better model financial variables such as market indices and credit spreads. To achieve this, we plan to develop different machine learning models that allow the modelling of these different financial variables based on cutting-edge sentiment analysis and other advanced natural language processing algorithms applied to public online texts. This internship project with its industry partner, a financial risk management software provider, plans to investigate comprehensive financial modelling techniques that incorporate natural language processing techniques such as sentiment analysis. The intern will work specifically on developing and tuning different machine learning models to better explain and predict the dynamics of financial variables interested.

Faculty Supervisor:

Roy Kwon

Student:

Partner:

SS&C Technologies

Discipline:

Engineering

Sector:

Professional, scientific and technical services

University:

University of Toronto

Program: