Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.
Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.
Author ORCID Identifier
https://orcid.org/0000-0003-1022-4721
AccessType
Open Access Dissertation
Document Type
dissertation
Degree Name
Doctor of Philosophy (PhD)
Degree Program
Computer Science
Year Degree Awarded
2024
Month Degree Awarded
February
First Advisor
Hong Yu
Second Advisor
Fei Liu
Subject Categories
Computational Engineering
Abstract
A major challenge in information understanding stems from the diverse nature of the audience, where individuals possess varying preferences, experiences, educational and cultural backgrounds. Consequently, adopting a one-size-fits-all approach to provide information may prove suboptimal. While prior research has predominantly focused on delivering pre-existing content to users with potential interests, this thesis explores generative language models for personalized information understanding. By harnessing the potential of generative language models, our objective is to generate novel personalize content for individual users. As a result, users from diverse backgrounds can be provided with content that are tailored for their need and better aligns with their interests. The crux of this research hinges on addressing the following two aspects: 1. Personalized Content: How to harness user profiles to create tailored content for individual users; 2. Effective Communication: How to engage with users in order to proficiently convey information. For the first aspect, i.e. personalized content, we explored personalized news headline generation. By analyzing users' reading history, our proposed framework identifies perspectives that users are interested in, which can further guide generating news headlines that are attractive to users. For the second aspect, i.e. effective communication, we developed personalized reading assistive agent, which assist users understand complex information in news article or academic documents through conversations. Compared to reading, obtaining information through conversations is more interactive and requires shorter attention span. We further incorporate the above aspects in personalized information systems in a real-life scenario, i.e. patient education. Specifically, we propose a novel after-visit summaries (AVS) writing assistant. After-visit summaries notes are documents given to patients to help them understand their clinical visits and disease self-management. Our approach not only automatically generates AVS drafts, but also detects potential errors in the generated drafts, allowing physicians to revise and produce AVS notes with higher efficiency and accuracy. Moreover, we present PaniniQA, a patient-centric interactive question answering system designed to help patients understand their discharge instructions. PaniniQA first identifies important clinical content from patients’ discharge instructions and then formulates personalized educational questions for distinctive patients. In addition, PaniniQA is also equipped with answer verification functionality to provide timely feedback to correct patients’ misunderstandings. Overall, we aspire to contribute to the advancement of information dissemination techniques, promoting a more inclusive and effective means of communication in our information-driven world.
DOI
https://doi.org/10.7275/36214447
Recommended Citation
Cai, Pengshan, "Generative Language Models for Personalized Information Understanding" (2024). Doctoral Dissertations. 3036.
https://doi.org/10.7275/36214447
https://scholarworks.umass.edu/dissertations_2/3036
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.