Jove
Visualize
Contact Us

Related Experiment Videos

Using context to build semantics.

Peter J Kwantes1

  • 1Defence R&D Canada, Toronto, Ontario, Canada. peter.kwantes@drdc-rddc.gc.ca

Psychonomic Bulletin & Review
|February 2, 2006
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

Comparing methods for single paragraph similarity analysis.

Topics in cognitive science·2014
Same author

A sequential sampling account of response bias and speed-accuracy tradeoffs in a conflict detection task.

Journal of experimental psychology. Applied·2013
Same author

Right inferior frontal gyrus activation as a neural marker of successful lying.

Frontiers in human neuroscience·2013
Same author

Adaptive decision making in a dynamic environment: a test of a sequential sampling model of relative judgment.

Journal of experimental psychology. Applied·2013
Same author

Item order matters in a function learning task.

Canadian journal of experimental psychology = Revue canadienne de psychologie experimentale·2012
Same author

An evidence accumulation model for conflict detection performance in a simulated air traffic control task.

Human factors·2009
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

A new memory model creates word representations similar to Latent Semantic Analysis (LSA) by using episodic memory retrieval instead of dimension reduction. This approach captures semantic similarity through memory recall mechanisms.

Area of Science:

  • Cognitive Science
  • Natural Language Processing
  • Computational Linguistics

Background:

  • Latent Semantic Analysis (LSA) models word knowledge using dimension reduction on co-occurrence data.
  • LSA identifies semantic similarity through higher-order associations derived from singular value decomposition.
  • Existing models primarily rely on statistical methods for semantic representation.

Purpose of the Study:

  • To introduce a novel memory model for generating semantic word representations.
  • To explore an alternative approach to LSA's dimension reduction technique.
  • To investigate the role of episodic memory in word meaning representation.

Main Methods:

  • A retrieval mechanism based on episodic memory principles was employed.

Related Experiment Videos

  • The model constructs semantic representations without applying dimension reduction.
  • Word associations were built through a process analogous to memory recall.

    • Main Results:

    • The developed memory model produces semantic representations comparable to LSA.
    • The model successfully captures higher-order associations relevant to semantic similarity.
    • Episodic memory retrieval proves effective for generating condensed word representations.

    Conclusions:

    • Episodic memory retrieval offers a viable alternative to dimension reduction for semantic analysis.
    • This memory-based approach provides a new perspective on computational models of word meaning.
    • The findings suggest a link between episodic memory and semantic knowledge representation.