Connotation Frames: A Data-Driven Investigation

TL;DR

This paper introduces connotation frames as a formalism to capture subtle implied sentiments and presuppositions in language, demonstrating their extraction from data and potential for bias analysis in media.

Contribution

It presents a novel formalism for representing connotative meanings and models for predicting these frames from language data, validated through crowdsourcing and empirical experiments.

Findings

Connotation frames can be reliably obtained via crowdsourcing.

Models accurately predict connotation frames from distributional representations.

Connotation frames reveal subtle biases in online news media.

Abstract

Through a particular choice of a predicate (e.g., "x violated y"), a writer can subtly connote a range of implied sentiments and presupposed facts about the entities x and y: (1) writer's perspective: projecting x as an "antagonist"and y as a "victim", (2) entities' perspective: y probably dislikes x, (3) effect: something bad happened to y, (4) value: y is something valuable, and (5) mental state: y is distressed by the event. We introduce connotation frames as a representation formalism to organize these rich dimensions of connotation using typed relations. First, we investigate the feasibility of obtaining connotative labels through crowdsourcing experiments. We then present models for predicting the connotation frames of verb predicates based on their distributional word representations and the interplay between different types of connotative relations. Empirical results confirm that connotation frames can be induced from various data sources that reflect how people use language and give rise to the connotative meanings. We conclude with analytical results that show the potential use of connotation frames for analyzing subtle biases in online news media.