Extracting semantic representations from word co-occurrence statistics: A computational study
- 1 August 2007
- journal article
- research article
- Published by Springer Science and Business Media LLC in Behavior Research Methods
- Vol. 39 (3), 510-526
- https://doi.org/10.3758/bf03193020
Abstract
The idea that at least some aspects of word meaning can be induced from patterns of word co-occurrence is becoming increasingly popular. However, there is less agreement about the precise computations involved, and the appropriate tests to distinguish between the various possibilities. It is important that the effect of the relevant design choices and parameter values are understood if psychological models using these methods are to be reliably evaluated and compared. In this article, we present a systematic exploration of the principal computational possibilities for formulating and validating representations of word meanings from word co-occurrence statistics. We find that, once we have identified the best procedures, a very simple approach is surprisingly successful and robust over a range of psychologically relevant evaluation measures.Keywords
This publication has 24 references indexed in Scilit:
- Metaphor Comprehension: What Makes a Metaphor Difficult to Understand?Metaphor and Symbol, 2002
- Unsupervised Learning by Probabilistic Latent Semantic AnalysisMachine Learning, 2001
- Symbol Grounding and Meaning: A Comparison of High-Dimensional and Embodied Theories of MeaningJournal of Memory and Language, 2000
- Theory and Operational Definitions in Computational Memory Models: A Response to Glenberg and RobertsonJournal of Memory and Language, 2000
- Learning to Segment Speech Using Multiple Cues: A Connectionist ModelLanguage and Cognitive Processes, 1998
- A solution to Plato's problem: The latent semantic analysis theory of acquisition, induction, and representation of knowledge.Psychological Review, 1997
- Self-Organizing MapsPublished by Springer Science and Business Media LLC ,1997
- Indexing by latent semantic analysisJournal of the American Society for Information Science, 1990
- The symbol grounding problemPhysica D: Nonlinear Phenomena, 1990
- Category norms of verbal items in 56 categories A replication and extension of the Connecticut category norms.Journal of Experimental Psychology, 1969