Literacy refers to the ability to read and write to acquire and convey information. High literacy has positive impacts on both individuals and the entire society. However, 21% of adults in the US were illiterate, including 4.1% classified as "functionally illiterate" [1]. This thesis explores natural language processing (NLP) technologies to help human users achieve better literacy by assisting them in reading and writing. We explore writing assistive technologies in both the clinical and literature domains:
For reading assistive technology, we explore a novel method for users to interactively acquire information, i.e. a user does not gain information through reading a passage but through conversations with an AI-powered chatbot who reads the passage. The chatbot actively leads the conversation and interactively addresses the user's questions, making the information acquisition process more engaging and effective.
As proposed work, we aim to apply our information acquisition-oriented chatbot to patient education, i.e. helping patients understand and memorize their health situation and self-management instructions better through conversations with chatbots.
Advisor: Hong Yu