Event extraction from Twitter using Non-Parametric Bayesian Mixture Model with Word Embeddings.
Deyu ZhouXuan ZhangYulan HePublished in: EACL (1) (2017)
Keyphrases
- mixture model
- event extraction
- border security
- online news
- maximum likelihood
- probability density function
- open domain
- em algorithm
- information extraction
- gaussian mixture model
- probabilistic model
- gaussian processes
- cross lingual
- density estimation
- news articles
- generative model
- language model
- n gram
- information fusion
- expectation maximization
- model selection
- co occurrence
- information access
- generalized em algorithm
- vector space
- topic detection
- unknown words
- gaussian process
- news stories
- word segmentation
- topic models
- unsupervised learning
- low dimensional
- latent topics
- natural language processing
- knowledge discovery
- reinforcement learning
- feature selection
- computer vision