Multilingual Topic Models for Unaligned Text

2012/05/09 by Boyd-Graber, Jordan, Blei, David · 3 citations
#Computation and Language (cs.CL) #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.1205.2657

Abstract

We develop the multilingual topic model for unaligned text (MuTo), a probabilistic model of text that is designed to analyze corpora composed of documents in two languages. From these documents, MuTo uses stochastic EM to simultaneously discover both a matching between the languages and multilingual latent topics. We demonstrate that MuTo is able to find shared topics on real-world multilingual corpora, successfully pairing related documents across languages. MuTo provides a new framework for creating multilingual topic models without needing carefully curated parallel corpora and allows applications built using the topic model formalism to be applied to a much wider class of corpora.

Cited by

• XTRA: Cross-Lingual Topic Modeling with Topic and Representation Alignments
• TMT: A Simple Way to Translate Topic Models Using Dictionaries
• How Historical Context Shapes the Construction of Field-Level Frames: A comparative study of France and Germany’s shale gas debates

Related

• Learning Multilingual Topics from Incomparable Corpus
• Bilingual Topic Models for Comparable Corpora
• Multilingual and Multimodal Topic Modelling with Pretrained Embeddings
• TMT: A Simple Way to Translate Topic Models Using Dictionaries
• A Bayesian Multilingual Document Model for Zero-shot Topic Identification and Discovery