Overweight and Obesity Stigma within the Saudi Community: what does ‘Sentiment Analysis’ Technique tell us?, Dr Dalia El Kheir, Reema Alghamdi, Raghad Badoghaish, Prof. Yahya AlMurtadha, Dr Abdelrahman Elfaki
Dr Dalia Yahia M. El Kheir
Consultant, Department of Family and Community Medicine
Imam Abdulrahman bin Faisal University
Dammam
Saudi Arabia
ORCID: 0000-0001-6033-1045
Reema Jamaan Alghamdi
Student, College of Medicine
Imam Abdulrahman bin Faisal University
Dammam
Saudi Arabia
ORCID: 0009-0006-6455-0468
Raghad Adel Badoghaish
Student, College of Medicine
Imam Abdulrahman bin Faisal University
Dammam
Saudi Arabia
ORCID: 0009-0006-1697-1655
Professor Yahya AlMurtadha
Professor, Computer Science Institution
University of Tabuk
Tabuk
Saudi Arabia
ORCID: 0000-0003-2171-5470
Dr Abdelrahman Osman Elfaki
Associate Professor, Computer Science Institution
University of Tabuk
Tabuk
Saudi Arabia
ORCID: 0000-0002-8881-0504
Paper Type: Research
DOI: 10.47556/B.OUTLOOK2025.23.12
Received: 14 October 2025 / Revised: 29 October 2025 / Accepted: 17 December 2025 / Published: 30 December 2025
Purpose: This study examines public sentiment in Saudi Arabia towards overweight and obesity, focusing on the impact of the COVID-19 pandemic. It explores how social media discourse reinforces weight stigma and affects public perception and behaviour.
Design/Methodology/Approach: Using Twitter API v2, Arabic-language posts geolocated to Saudi Arabia were collected and filtered by relevant keywords. Sentiment analysis was performed using natural language processing and machine learning.
Findings: Over 96% of tweets referencing overweight and obesity expressed negative sentiment. Posts linking these terms to COVID-19 or weight gain were similarly unfavourable. Digital stigma may undermine health outcomes and compliance.
Value/Originality: The study shows how Artificial Intelligence (AI)-driven sentiment analysis can expose real-time health biases to inform stigma-sensitive policy.
Research Limitations/Implications: Dialectal variation and linguistic complexity may affect sentiment detection, although dataset size ensured robust classification.
Practical Implications: Sentiment analysis supports health surveillance by identifying stigmatising narratives and guiding culturally responsive interventions.
Keywords: Sentiment Analysis; Artificial Intelligence Model (AI); Natural Language Processing; Social Media; Twitter; Obesity; Overweight; COVID-19.
