Meta-complementing the semantics of short texts in neural topic models

• Main Authors: ZHANG, Ce, LAUW, Hady Wirawan
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2022
• Subjects: Topic models; short documents; document connectivity; improved topic quality; Databases and Information Systems; Numerical Analysis and Scientific Computing; Theory and Algorithms
• Online Access: https://ink.library.smu.edu.sg/sis_research/7609; https://ink.library.smu.edu.sg/context/sis_research/article/8612/viewcontent/neurips22.pdf
• Institution: Singapore Management University

Topic models infer latent topic distributions based on observed word co-occurrences in a text corpus. While typically a corpus contains documents of variable lengths, most previous topic models treat documents of different lengths uniformly, assuming that each document is sufficiently informative. H...