Stereotypical gender actions can be extracted from web text

2011-09-01

J. Assoc. Inf. Sci. Technol.

AI Summary

This study investigates whether web-based text—particularly Twitter corpora—can effectively represent gender-stereotyped behavioral associations and align with human commonsense judgments. Methodologically, we propose a quantification framework for action-gender bias grounded in user gender metadata and pronoun/name heuristics, integrated with the Open Mind Common Sense knowledge base to systematically extract and annotate gender-associated actions. To our knowledge, this is the first cross-source validation of gendered behavioral stereotypes between web text and structured commonsense knowledge. We construct a high-quality gender-action dataset comprising 441 manually annotated and 21,442 automatically annotated instances. Experimental results show that our model achieves a Spearman correlation of 0.47 and an AUC of 0.76 against human-annotated gold standards, demonstrating that web text can robustly model and complement commonsense-level gendered behavioral stereotypes with high recall. This work establishes a novel paradigm for large-scale, dynamic modeling of gender cognition.

Technology Category

• Application Domains: Humanities & Computational Social Science
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Knowledge Representation and Reasoning: Common-Sense Reasoning

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

We extracted gender-specific actions from text corpora and Twitter, and compared them with stereotypical expectations of people. We used Open Mind Common Sense (OMCS), a common sense knowledge repository, to focus on actions that are pertinent to common sense and daily life of humans. We use the gender information of Twitter users and web-corpus-based pronoun/name gender heuristics to compute the gender bias of the actions. With high recall, we obtained a Spearman correlation of 0.47 between corpus-based predictions and a human gold standard, and an area under the ROC curve of 0.76 when predicting the polarity of the gold standard. We conclude that it is feasible to use natural text (and a Twitter-derived corpus in particular) in order to augment common sense repositories with the stereotypical gender expectations of actions. We also present a dataset of 441 common sense actions with human judges' ratings on whether the action is typically/slightly masculine/feminine (or neutral), and another larger dataset of 21,442 actions automatically rated by the methods we investigate in this study. © 2011 Wiley Periodicals, Inc.

Problem

• Extracting gender-specific actions from text corpora and Twitter
• Comparing extracted actions to stereotypical gender expectations
• Augmenting commonsense repositories with gender-biased action data

Innovation

• Extract gender-specific actions from text corpora
• Use OMCS for common sense action relevance
• Compute gender bias via Twitter user data