notesum.ai

Published at November 11

A Domain-Agnostic Neurosymbolic Approach for Big Social Data Analysis: Evaluating Mental Health Sentiment on Social Media during COVID-19

cs.AI

I.2.4; I.2.6; I.2.7; I.2.0

Released Date: November 11, 2024

Authors: Vedant Khandelwal¹, Manas Gaur², Ugur Kursuncu³, Valerie Shalin, Amit Sheth

Aff.: ¹AI Institute, University of South Carolina, Columbia, South Carolina; ²Department of Computer Science and Engineering, University of Maryland, Baltimore County, Maryland; ³Institute for Insight, Georgia State University, Atlanta, Georgia

Arxiv: http://arxiv.org/abs/2411.07163v1

Category	Model	Precision	Recall	F1-Score
Depression	NB	84.85 (-24%)	82.68 (-25%)	83.75 (-27%)
	RF	91.98 (-28%)	91.81 (-26%)	91.89 (-23%)
	BRF	92.32 (-27%)	92.43 (-24%)	92.37 (-29%)
	BSRF	94.12 (-29%)	93.02 (-22%)	93.57 (-28%)
Addiction	NB	82.74 (-26%)	80.46 (-21%)	81.58 (-25%)
	RF	90.02 (-22%)	90.36 (-20%)	90.19 (-23%)
	BRF	91.53 (-28%)	91.78 (-26%)	91.65 (-29%)
	BSRF	91.64 (-27%)	91.82 (-24%)	91.73 (-28%)
Anxiety	NB	82.53 (-25%)	81.87 (-24%)	82.20 (-22%)
	RF	90.76 (-23%)	92.78 (-28%)	91.76 (-21%)
	BRF	94.37 (-27%)	93.87 (-25%)	94.12 (-29%)
	BSRF	93.46 (-24%)	93.85 (-27%)	93.65 (-28%)