Deep topic modeling by multilayer bootstrap network and lasso

TL;DR

This paper introduces a novel deep topic modeling approach that combines multilayer bootstrap networks and Lasso, avoiding traditional assumptions and providing effective document dimension reduction and topic discovery.

Contribution

It is the first to apply multilayer bootstrap network and Lasso to unsupervised topic modeling, offering a model-free, efficient solution.

Findings

Outperforms five representative topic models on 20-newsgroups and TDT2 datasets.

Effectively reduces document dimensions and discovers meaningful topics.

Demonstrates robustness without relying on strong model or data assumptions.

Abstract

Topic modeling is widely studied for the dimension reduction and analysis of documents. However, it is formulated as a difficult optimization problem. Current approximate solutions also suffer from inaccurate model- or data-assumptions. To deal with the above problems, we propose a polynomial-time deep topic model with no model and data assumptions. Specifically, we first apply multilayer bootstrap network (MBN), which is an unsupervised deep model, to reduce the dimension of documents, and then use the low-dimensional data representations or their clustering results as the target of supervised Lasso for topic word discovery. To our knowledge, this is the first time that MBN and Lasso are applied to unsupervised topic modeling. Experimental comparison results with five representative topic models on the 20-newsgroups and TDT2 corpora illustrate the effectiveness of the proposed algorithm.