EMP-EVAL: A Framework for Measuring Empathy in Open Domain Dialogues

01/29/2023
by   Bushra Amjad, et al.
1

Measuring empathy in conversation can be challenging, as empathy is a complex and multifaceted psychological construct that involves both cognitive and emotional components. Human evaluations can be subjective, leading to inconsistent results. Therefore, there is a need for an automatic method for measuring empathy that reduces the need for human evaluations. In this paper, we proposed a novel approach EMP-EVAL, a simple yet effective automatic empathy evaluation method. The proposed technique takes the influence of Emotion, Cognitive and Emotional empathy. To the best knowledge, our work is the first to systematically measure empathy without the human-annotated provided scores. Experimental results demonstrate that our metrics can correlate with human preference, achieving comparable results with human judgments.

READ FULL TEXT

page 1

page 6

research
12/28/2022

Emotion in Cognitive Architecture: Emergent Properties from Interactions with Human Emotion

This document presents endeavors to represent emotion in a computational...
research
09/07/2021

POSSCORE: A Simple Yet Effective Evaluation of Conversational Search with Part of Speech Labelling

Conversational search systems, such as Google Assistant and Microsoft Co...
research
10/06/2020

Emotional Design

Emotional design has been well recognized in the domain of human factors...
research
07/06/2021

Comparing PCG metrics with Human Evaluation in Minecraft Settlement Generation

There are a range of metrics that can be applied to the artifacts produc...
research
05/14/2023

A Cognitive Stimulation Dialogue System with Multi-source Knowledge Fusion for Elders with Cognitive Impairment

When communicating with elders with cognitive impairment, cognitive stim...
research
08/16/2021

Cognitive AI Systems Contribute to Improving Creativity Modeling and Measuring Tools

Cognitive science and cognitive psychology have long used creativity tes...
research
05/20/2022

Measuring algorithmic interpretability: A human-learning-based framework and the corresponding cognitive complexity score

Algorithmic interpretability is necessary to build trust, ensure fairnes...

Please sign up or login with your details

Forgot password? Click here to reset