Discovering Evolutionary Theme Patterns from Text ∗ CS 598

Temporal Text Mining (TTM) is concerned with discovering temporal patterns in text information collected over time. Since most text information bears some time stamps, TTM has many applications in multiple domains, such as summarizing events in news articles and revealing research trends in scientific literature. In this paper, we study a particular TTM task – discovering and summarizing the evolutionary patterns of themes in a text stream. The evolutionary patterns of a theme are divided into content evolution and strength evolution. We define the problem of discovering evolutionary theme patterns on both aspects and present general probabilistic methods for (1) discovering latent themes from text; (2) constructing an evolution graph of themes; and (3) analyzing life cycles of themes. Evaluation of the proposed methods on three different data collections (i.e., one news articles collection and two literature collections) shows that the proposed methods can discover interesting evolutionary theme patterns effectively.