TY - BOOK TI - Topics over time: a non-Markov continuous-time model of topical trends AU - McCallum, Andrew AU - Wang, Xuerui AB - This paper presents an LDA-style topic model that captures not only the low-dimensional structure of data, but also how the structure changes over time. Unlike other recent work that relies on Markov assumptions or discretization of time, here each topic is associated with a continuous distribution over timestamps, and for each generated document, the mixture distribution over topics is influenced by both word co-occurrences and the document's timestamp. Thus, the meaning of a particular topic can be relied upon as constant, but the topics' occurrence and correlations change significantly over time. The presented results are on nine months of personal email, 17 years of NIPS research papers and over 200 years of presidential state-of-the-union addresses, showing improved topics, better timestamp prediction, and interpretable trends. DA - 2006/// PY - 2006 LA - en ST - Topics over time UR - /paper/Topics-over-time%3A-a-non-Markov-continuous-time-of-Wang-McCallum/7f8abf25ca24b48450b4e535f41e2b8a87df73f5 Y2 - 2019/01/13/00:00:00 KW - Topic model algorithm KW - Topic model longitudinal over time KW - Topic modeling ER -