Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

New trends in natural language processing: statistical natural language processing

M Marcus1

  • 1Department of Computer and Information Science, University of Pennsylvania, Philadelphia 19104-6389, USA.

Proceedings of the National Academy of Sciences of the United States of America
|October 24, 1995
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Antibiotic use in a co-infection of respiratory syncytial virus and pathogenic bacteria in children in a resource-limited setting in northeast Peninsular Malaysia.

Tropical biomedicine·2024
Same author

Developing Children's Oral Health Assessment Toolkits Using Machine Learning Algorithm.

JDR clinical and translational research·2019
Same author

Child and Parent Demographic Characteristics and Oral Health Perceptions Associated with Clinically Measured Oral Health.

JDR clinical and translational research·2019
Same author

Magnetic micro-device for manipulating PC12 cell migration and organization.

Lab on a chip·2015
Same author

NGF-conjugated iron oxide nanoparticles promote differentiation and outgrowth of PC12 cells.

Nanoscale·2014
Same author

Association between pubertal development and depressive symptoms in girls from a UK cohort.

Psychological medicine·2012
Same journal

The TaMYB55-TaSnRK1α1-TabZIP9 module confers heat stress tolerance in wheat.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Superstatistics approach to turbulent circulation fluctuations.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

A molecular timescale for evolution of cobamide biosynthesis.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Pierre Chambon, a pioneer of molecular biology and gene regulation in eukaryotes.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Granulosa cell glycogen fuels the avascular corpus luteum.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Synthetic essentiality of TRAIL/TNFSF10 in VHL-deficient renal cell carcinoma.

Proceedings of the National Academy of Sciences of the United States of America·2026
See all related articles

Natural Language Processing (NLP) research now combines symbolic methods with empirical, probabilistic techniques. This shift, enabled by annotated linguistic databases, significantly improves NLP system performance.

Area of Science:

  • Computational Linguistics
  • Natural Language Processing (NLP)

Background:

  • Traditional computational linguistics relied heavily on symbolic methods.
  • Recent advancements show a shift towards hybrid approaches in NLP.
  • Availability of annotated linguistic databases enables new empirical methods.

Purpose of the Study:

  • To survey recent trends in Natural Language Processing (NLP).
  • To highlight the shift from symbolic to hybrid methods in NLP.
  • To focus on progress in part-of-speech tagging, stochastic parsing, and lexical semantics.

Main Methods:

  • Combining empirical corpus-based methods with traditional symbolic techniques.
  • Utilizing probabilistic and information-theoretic approaches.
  • Leveraging annotated linguistic databases for natural language text corpora.

Related Experiment Videos

Main Results:

  • Dramatic improvements in the performance of various NLP systems.
  • Significant progress in part-of-speech tagging accuracy.
  • Advancements in stochastic parsing and lexical semantics.

Conclusions:

  • Hybrid methods represent a significant advancement in NLP.
  • Continued improvements in NLP system performance are expected.
  • The integration of empirical and symbolic approaches is key to future NLP progress.