This project is a continuation of last semester's work with Intuit, which centered on working with large amounts of unstructured anonymized user data, such as calendar events, emails, finances, etc. to reconstruct a detailed timeline representing the evolution of a user's life over time, event by event. This allows Intuit to better understand the nature of its customers and build more accurate customer profiles which directly influence the quality of their products. This semester, the team will be tackling sub-problems towards this goal. Specifically, they'll be working on identifying, extracting, and computationally understanding event-based text in emails, and in particular extract information from these events that's relevant to taxes. Towards this goal, the project involves working with computational linguistics algorithms such as named entity and theme recognition, and will be taking a deep learning approach to the text analysis.
No Team Members