• Publications
  • Influence
Reading Tea Leaves: How Humans Interpret Topic Models
TLDR
We present a method for measuring the interpretatability of a topic model that captures aspects of the model that are undetected by previous metrics. Expand
  • 1,338
  • 108
  • PDF
A Neural Network for Factoid Question Answering over Paragraphs
TLDR
We introduce a recursive neural network (rnn) model that can reason over such input by modeling textual compositionality. Expand
  • 278
  • 16
  • PDF
Adding dense, weighted connections to WordNet
TLDR
WORDNET, a ubiquitous tool for natural language processing, suffers from sparsity of connections between its component concepts (synsets). Expand
  • 139
  • 16
  • PDF
Maximum Likelihood
In statistics, maximum-likelihood estimation (MLE) is a method of estimating the parameters of a statistical model. When applied to a data set and given a statistical model, maximum-likelihoodExpand
  • 89
  • 14
  • PDF
Syntactic Topic Models
TLDR
We develop the syntactic topic model (STM), a nonparametric Bayesian model of parsed documents. Expand
  • 180
  • 11
  • PDF
Political Ideology Detection Using Recursive Neural Networks
TLDR
We apply recursive neural networks to political ideology detection, a problem where previous work relies heavily on bag-of-words models and hand-designed lexica. Expand
  • 166
  • 8
  • PDF
Multilingual Topic Models for Unaligned Text
TLDR
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. Expand
  • 152
  • 8
  • PDF
The Amazing Mysteries of the Gutter: Drawing Inferences Between Panels in Comic Book Narratives
TLDR
We present the COMICS dataset, which contains over 1.2 million panels drawn from almost 4,000 publicly-available comic books published during the “Golden Age” of American comics. Expand
  • 38
  • 8
  • PDF
Feuding Families and Former Friends: Unsupervised Learning for Dynamic Fictional Relationships
TLDR
We formalize the task of unsupervised relationship modeling, which involves learning a set of relationship descriptors as well as a trajectory over these descriptors for each relationship in an input dataset. Expand
  • 78
  • 6
  • PDF
Modeling topic control to detect influence in conversations using nonparametric topic models
TLDR
We introduce SITS—Speaker Identity for Topic Segmentation, a nonparametric hierarchical Bayesian model that is capable of discovering (1) the topics used in a set of conversations, (2) how these topics are shared across conversations, and (3) when these topics change during conversations. Expand
  • 37
  • 5
  • PDF