2012 · 17 citations · 18 references
ComputingComputer ScienceComputational LinguisticsNatural Language ProcessingText AnalyticsHidden Markov ModelsLanguage ModelingLanguage ModelsPart-of-speech TaggingSyntactic AnalysisTopic ModelingBayesian Topic ModelsNoisy Text Part-of-speech TaggingSemantic InformationSide Information UtilizationUnsupervised Part-of-speech TaggingUnsupervised Topic Modeling
Unsupervised part-of-speech (POS) tagging has recently been shown to greatly benefit from Bayesian approaches where HMM parameters are integrated out, leading to significant increases in tagging accuracy. These improvements in unsupervised methods are important especially in specialized social media domains such as Twitter where little training data is available. Here, we take the Bayesian approach one step further by integrating semantic information from an LDA-like topic model with an HMM. Specifically, we present Part-of-Speech LDA (POSLDA), a syntactically and semantically consistent generative probabilistic model. This model discovers POS specific topics from an unlabelled corpus. We show that this model consistently achieves improvements in unsupervised POS tagging and language modeling over the Bayesian HMM approach with varying amounts of side information in the noisy and esoteric domain of Twitter. 1
18
Reading Tea Leaves: How Humans Interpret Topic Models
Jonathan Chang, Sean Gerrish, Chong Wang et al. · 2009 · 1.9K citations
Comparing clusterings—an information based distance
Marina Meilă · Journal of Multivariate Analysis · 2006 · 1.6K citations
Part-of-Speech Tagging for Twitter: Annotation, Features, and Experiments
Kevin Gimpel, Nathan Schneider, Brendan O’Connor et al. · Figshare · 2018 · 830 citations · Full text
Model selection and model averaging
Gerda Claeskens, Nils Lid Hjort · RePEc: Research Papers in Economics · 2008 · 773 citations