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

A Customized Attention-Based Long Short-Term Memory Network for Distant Supervised Relation Extraction.

Dengchao He1, Hongjun Zhang2, Wenning Hao3

  • 1College of Command Information System, PLA University of Science and Technology, Nan Jing, 210007, P.R.C. hdchao1989@163.com.

Neural Computation
|April 15, 2017
PubMed
Summary

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

Synergistic anion-cation descriptor for bidirectional electrocatalyst in Li-CO<sub>2</sub> battery.

Science advances·2026
Same author

Mechanical characteristics and solidification mechanism of sodium silicate activated fly ash cement composite loess based fluidized stabilized materials.

Scientific reports·2026
Same author

LF3 exerts anti-osteosarcoma effects through Wnt signaling inhibition-coupled ferroptosis induction via <i>HO-1/ACSL4</i> axis.

Frontiers in oncology·2026
Same author

Effects of High-Intensity Interval Training on Body Composition, Metabolic Health, and Cardiorespiratory Fitness in Overweight or Obese Children and Adolescents: A Systematic Review and Meta-Analysis.

Metabolites·2026
Same author

Response characteristic of bolt connection stress to jacking effect in prefabricated assembled utility tunnel in Xiong'an New Area, China.

Scientific reports·2026
Same author

Fraction Conversion Models Based on Ultrasound Attenuation Coefficient for Assessing Liver Steatosis.

Diagnostics (Basel, Switzerland)·2026
This summary is machine-generated.

This study introduces an attention-based network to improve relation extraction using distant supervision. The model effectively handles noisy data and reduces reliance on manual feature engineering for better performance.

Area of Science:

  • Natural Language Processing
  • Machine Learning
  • Artificial Intelligence

Background:

  • Distant supervision is a common method for generating labeled data in relation extraction, but it often introduces false positives.
  • Traditional feature-based methods in relation extraction rely on manual feature engineering, which can limit performance.
  • Existing approaches struggle with the inherent noise and data representation challenges in large-scale, automatically generated datasets.

Purpose of the Study:

  • To develop a novel approach for relation extraction that overcomes the limitations of traditional distant supervision.
  • To improve data representation and mitigate the impact of false-positive data in automatically labeled corpora.
  • To eliminate the need for manually designed features in distant supervised relation extraction models.

Related Experiment Videos

Main Methods:

  • A customized attention-based long short-term memory (LSTM) network was proposed.
  • Word-level attention was employed for enhanced data representation without manual feature engineering.
  • Instance-level attention was utilized to address the issue of false-positive data inherent in distant supervision.

Main Results:

  • The proposed attention-based LSTM network demonstrated effective performance in relation extraction.
  • The approach achieved superior results compared to traditional feature-based distant supervised methods.
  • Both word-level and instance-level attention mechanisms contributed to improved data representation and noise reduction.

Conclusions:

  • The customized attention-based LSTM network is an effective solution for distant supervised relation extraction.
  • The proposed method successfully addresses the challenges of noisy data and manual feature engineering.
  • This approach offers a promising direction for improving the accuracy and efficiency of relation extraction systems.