Multiscale topic tomography
Authors:
Open Access
- 12 August 2007
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 520-529
Abstract
Modeling the evolution of topics with time is of great value in automatic summarization and analysis of large document collections. In this work, we propose a new probabilistic graphical model to address this issue. The new model, which we call the Multiscale Topic Tomography Model (MTTM), employs non-homogeneous Poisson processes to model gen- eration of word-counts. The evolution of topics is modeled through a multi-scale analysis using Haar wavelets. One of the new features of the model is its modeling the evolution of topics at various time-scales of resolution, allowing the user to zoom in and out of the time-scales. Our experiments on Science data using the new model uncovers some interest- ing patterns in topics. The new model is also comparable to LDA in predicting unseen data as demonstrated by our perplexity experiments.Keywords
This publication has 5 references indexed in Scilit:
- Topics over time2006
- Dynamic topic models2006
- GaP2004
- A statistical multiscale framework for Poisson inverse problemsIEEE Transactions on Information Theory, 2000
- An Introduction to Variational Methods for Graphical ModelsMachine Learning, 1999