Continuous-Time Infinite Dynamic Topic Models: The Dim Sum Process for Simultaneous Topic Enumeration and Formation

Continuous-Time Infinite Dynamic Topic Models: The Dim Sum Process for Simultaneous Topic Enumeration and Formation

Wesam Elshamy (Kansas State University, USA) and William H. Hsu (Kansas State University, USA)
DOI: 10.4018/978-1-4666-5063-3.ch009
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Topic models are probabilistic models for discovering topical themes in collections of documents. These models provide us with the means of organizing what would otherwise be unstructured collections. The first wave of topic models developed was able to discover the prevailing topics in a big collection of documents spanning a period of time. These time-invariant models were not capable of modeling 1) the time varying number of topics they discover and 2) the time changing structure of these topics. Few models were developed to address these two deficiencies. The online-hierarchical Dirichlet process models the documents with a time varying number of topics, and the continuous-time dynamic topic model evolves topic structure in continuous-time. In this chapter, the authors present the continuous-time infinite dynamic topic model that combines the advantages of these two models. It is a probabilistic topic model that changes the number of topics and topic structure over continuous-time.
Chapter Preview
Top

1. Dynamic Topic Modeling And Event Stream Mining

In this chapter, we discuss the problem of simultaneous topic enumeration and tracking (STEF): that of maintaining both the number of topics and a parametric representation of their changing semantics as estimated from a dynamically changing document collection, or corpus. We develop a continuous-time dynamic topic model for addressing the STEF problem. We extend the existing Bayesian representation of evolving topic models with two temporal components: the word distribution per topic, and the number of topics. Both evolve in continuous time.

Complete Chapter List

Search this Book:
Reset