Neural labeled LDA: a topic model for semi-supervised document classification

Abstract

Recently, some statistical topic modeling approaches based on LDA have been applied in the field of supervised document classification, where the model generation procedure incorporates prior knowledge to improve the classification performance. However, these customizations of topic modeling are limited by the cumbersome derivation of a specific inference algorithm for each modification. In this paper, we propose a new supervised topic modeling approach for document classification problems, Neural Labeled LDA (NL-LDA), which builds on the VAE framework, and designs a special generative network to incorporate prior information. The proposed model can support semi-supervised learning based on the manifold assumption and low-density assumption. Meanwhile, NL-LDA has a consistent and concise inference method while semi-supervised learning and predicting. Quantitative experimental results demonstrate our model has outstanding performance on supervised document classification relative to the compared approaches, including traditional statistical and neural topic models. Specially, the proposed model can support both single-label and multi-label document classification. The proposed NL-LDA performs significantly well on semi-supervised classification, especially under a small amount of labeled data. Further comparisons with related works also indicate our model is competitive with state-of-the-art topic modeling approaches on semi-supervised classification.

Hot Topic Recognition of Health Rumors Based on Anti-Rumor Articles on the WeChat Official Account Platform: Topic Modeling

“…By providing text input and by setting the desired number of topics, LDA automatically generates a set of topics, assigns words to topics, and assigns a topic ratio to each document. The LDA topic model is effective in transforming text categorization problems into machine-understandable linguistic problems with improved accuracy [ 34 , 35 ]. By using Gibbs sampling—a method for estimating the marginal distributions of the variables of interest—the LDA topic model can determine the topics in the data pool.…”

Automated Extraction of Rail Point Clouds by Multi-Scale Dimensional Features From MLS Data

“…Liner Discriminant Analysis(LDA), also named Fisher Liner Discriminant, is a classic algorithm for model identification [41]- [43]. The idea is to project high-dimensional data into low-dimensional so that it has the smallest intra-class distance and the largest inter-class distance in the optimal discriminant vector space.…”

Multisubject Analysis and Classification of Books and Book Collections, Based on a Subject Term Vocabulary and the Latent Dirichlet Allocation

“…TL-LDA has proven effective in both single-and multi-label classification experiments. Neural labeled LDA [16] is another method that supports both supervised and semi-supervised document classification. It is built on SLDA [17] and the semisupervised document classification is based on the manifold assumption and low-density assumption.…”