In the field of alternative investment management, big data presents challenges and opportunities, necessitating a shift in how firms approach information management. Our team collaborated with an alternative investment firm to reduce the amount of time analysts review potential and historical investment deal data. To achieve this goal, we executed a two-part project: the first part involved constructing a Power BI dashboard of key metrics for their Residential Mortgage-Backed Securities desk using Azure Data Lake, Python, Power BI, and PostgreSQL; the second part utilized the company’s proprietary Chat-GPT system and prompt engineering to develop a deal appraisal model using Python, Databricks, Azure Cognitive Search/AI Document Intelligence, and Amazon Textract.

• This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator

• Liao, Maya
• Shemesh, Nathan
• Amicarella, Dante
• LaRusso, Sarah

Subject

• Innovation
• Industry
• Project-Based Learning
• Computing

Publisher

• Worcester Polytechnic Institute

Identifier

• E-project-121323-122023
• 115084

Keyword

• ChatGPT OpenAI
• Data Science
• Large Language Model (LLM)
• Prompt Engineering
• Financial Deal Documents
• Azure Data Lake

Advisor

• Sarnie, Robert
• Blais, Marcel Y.
• Wong, Wilson
• Danke, Manasi

Year

• 2023

Sponsor

• Alternative Investment Company

UN Sustainable Development Goals

• Not Specified

Date created

• 12/13/2023

Resource type

• Major Qualifying Project

Major

• Mathematical Sciences
• Data Science
• Computer Science

Source

• E-project-121323-122023

Rights statement

• In Copyright

Last modified

• 2024-02-08