Extended overview of CLEF HIPE 2020: named entity processing on historical newspapers
This paper presents an extended overview of the first edition of HIPE (Identifying Historical
People, Places and other Entities), a pioneering shared task dedicated to the evaluation of …
People, Places and other Entities), a pioneering shared task dedicated to the evaluation of …
[HTML][HTML] De-identification of psychiatric intake records: Overview of 2016 CEGS N-GRID shared tasks Track 1
Abstract The 2016 CEGS N-GRID shared tasks for clinical records contained three tracks.
Track 1 focused on de-identification of a new corpus of 1000 psychiatric intake records. This …
Track 1 focused on de-identification of a new corpus of 1000 psychiatric intake records. This …
An introduction to conditional random fields
Many tasks involve predicting a large number of variables that depend on each other as well
as on other observed variables. Structured prediction methods are essentially a combination …
as on other observed variables. Structured prediction methods are essentially a combination …
[PDF][PDF] Efficient higher-order CRFs for morphological tagging
Training higher-order conditional random fields is prohibitive for huge tag sets. We present
an approximated conditional random field using coarse-to-fine decoding and early updating …
an approximated conditional random field using coarse-to-fine decoding and early updating …
Overview of CLEF HIPE 2020: Named entity recognition and linking on historical newspapers
This paper presents an overview of the first edition of HIPE (Identifying Historical People,
Places and other Entities), a pioneering shared task dedicated to the evaluation of named …
Places and other Entities), a pioneering shared task dedicated to the evaluation of named …
Natural language processing of radiology reports for the detection of thromboembolic diseases and clinically relevant incidental findings
Abstract Background Natural Language Processing (NLP) has been shown effective to
analyze the content of radiology reports and identify diagnosis or patient characteristics. We …
analyze the content of radiology reports and identify diagnosis or patient characteristics. We …
Curriculum-based transfer learning for an effective end-to-end spoken language understanding and domain portability
We present an end-to-end approach to extract semantic concepts directly from the speech
audio signal. To overcome the lack of data available for this spoken language …
audio signal. To overcome the lack of data available for this spoken language …
Non-uniform stochastic average gradient method for training conditional random fields
We apply stochastic average gradient (SAG) algorithms for training conditional random
fields (CRFs). We describe a practical implementation that uses structure in the CRF …
fields (CRFs). We describe a practical implementation that uses structure in the CRF …
[HTML][HTML] Stacked ensemble combined with fuzzy matching for biomedical named entity recognition of diseases
Abstract Biomedical Named Entity Recognition (Bio-NER) is the crucial initial step in the
information extraction process and a majorly focused research area in biomedical text …
information extraction process and a majorly focused research area in biomedical text …
Ensemble method–based extraction of medication and related information from clinical texts
Objective Accurate and complete information about medications and related information is
crucial for effective clinical decision support and precise health care. Recognition and …
crucial for effective clinical decision support and precise health care. Recognition and …