• Publications
  • Influence
Developing a Robust Part-of-Speech Tagger for Biomedical Text
TLDR
This paper presents a part-of-speech tagger which is specifically tuned for biomedical text. Expand
  • 503
  • 36
A Joint Many-Task Model: Growing a Neural Network for Multiple NLP Tasks
TLDR
We introduce a joint many-task model together with a strategy for successively growing its depth to solve increasingly complex tasks at successively deeper layers. Expand
  • 351
  • 36
  • PDF
Stochastic Gradient Descent Training for L1-regularized Log-linear Models with Cumulative Penalty
TLDR
Stochastic gradient descent (SGD) uses approximate gradients estimated from subsets of the training data and updates the parameters in an online fashion—the weights are updated much more frequently than batch training algorithms. Expand
  • 198
  • 25
  • PDF
Bidirectional Inference with the Easiest-First Strategy for Tagging Sequence Data
TLDR
This paper presents a bidirectional inference algorithm for sequence labeling problems such as part-of-speech tagging, named entity recognition and text chunking. Expand
  • 271
  • 22
  • PDF
Tree-to-Sequence Attentional Neural Machine Translation
TLDR
We propose a novel end-to-end syntactic NMT model to take advantage of syntactic infor- ar X iv :1 60 3. Expand
  • 167
  • 21
  • PDF
Improving Chinese Word Segmentation and POS Tagging with Semi-supervised Methods Using Large Auto-Analyzed Data
TLDR
This paper presents a simple yet effective semi-supervised method to improve Chinese word segmentation and POS tagging by incorporating large unlabeled data. Expand
  • 79
  • 16
  • PDF
Traction control of electric vehicle: basic experimental results using the test EV "UOT electric march"
The most distinct advantage of the electric vehicle is its quick and precise torque generation. However, most electric vehicles developed to date have not yet utilized this feature. In this paper,Expand
  • 255
  • 12
  • PDF
Learning with Lookahead: Can History-Based Models Rival Globally Optimized Models?
TLDR
This paper shows that the performance of history-based models can be significantly improved by performing lookahead in the state space when making each classification decision. Expand
  • 62
  • 12
  • PDF
Semantic Retrieval for the Accurate Identification of Relational Concepts in Massive Textbases
TLDR
This paper introduces a novel framework for the accurate retrieval of relational concepts from huge texts by applying a deep parser and a term recognizer. Expand
  • 119
  • 11
FACTA: a text search engine for finding associated biomedical concepts
TLDR
FACTA is a text search engine for MEDLINE abstracts, which is designed particularly to help users browse biomedical concepts (e.g. genes/proteins, diseases, enzymes and chemical compounds) appearing in the documents retrieved by the query. Expand
  • 169
  • 9
  • PDF