Natural Language Processing Researcher
Current• Applying Natural Language Processing and Latent Semantic Analysis to announcements from the Fed• Created a corpus of Federal Reserve speeches and minutes for natural language processing • Examining how the tone of transcripts and minutes of FOMC meetings changes over time and how those changes are related to economic indicators, using sentiment analysis • Generated the topic proportion for FOMC meetings and transcripts using machine learning techniques such as Latent Dirichlet allocation (LDA)• Comparing Fed transcripts to corresponding minutes to discover whether the minutes are an accurate summary of the transcripts, or if they are crafted to send specific messages to investors