Mining a Crowdsourced Dictionary to Understand Consistency and Preference in Word Meanings

Area of Science:

• Psychology
• Computational Linguistics
• Cognitive Science

Background:

• Big data methodologies are increasingly utilized in psychological research.
• Distributional models of semantics learn word meanings from large text corpora.
• Mega datasets of human behavior are valuable resources for linguistic studies.

Purpose of the Study:

• To investigate the consistency and preference in human word meanings.
• To combine distributional semantic models with large-scale behavioral data.
• To understand the cultural evolution of word meanings.

Main Methods:

• Mined data from an online, crowdsourced dictionary.
• Analyzed linguistic data using a distributional semantic model.
• Examined consistency and preferences in word meaning attribution.

Main Results:

• High consistency was observed in word meanings among different people, even for words not actively used.
• Language users exhibit strong preferences for specific word meanings.
• Community consensus rejects definitions that deviate from established conceptions.

Conclusions:

• Individual and community-level consistency exists in word meaning.
• Community preferences play a role in shaping and maintaining word meanings.
• This approach offers novel methods for studying lexical behavior and cultural evolution.