Learn More
OBJECTIVE To evaluate state-of-the-art unsupervised methods on the word sense disambiguation (WSD) task in the clinical domain. In particular, to compare graph-based approaches relying on a clinical knowledge base with bottom-up topic-modeling-based approaches. We investigate several enhancements to the topic-modeling techniques that use domain-specific(More)
We use Bayesian topic modeling techniques adapted to the task of unsupervised word sense induction on acronyms in clinical text and investigate (1) the amount of annotated data needed by such approaches to match the performance of the supervised sense disambiguation systems, and (2) feasibility of using an automatically collected silver standard for such(More)
Bayesian topic models have recently been shown to perform well in word sense induction (WSI) tasks. Such models have almost exclusively used bag-of-words features, and failed to attain improvement by including other feature types. In this paper, we investigate the impact of integrating syntactic and knowledge-based features and show that both parametric and(More)
— The stigma associated with mental health issues makes face-to-face discussions with family members, friends, or medical professionals difficult for many people. In contrast, the Internet, due to its ubiquity and global outreach, is increasingly becoming a popular medium for distressed individuals to anonymously relate experiences. In this paper, we(More)
  • 1