Detecting associations between dietary supplement intake and sentiments within mental disorder tweets

Wang, Yefeng; Zhao, Yunpeng; Zhang, Jianqiu; Bian, Jiang; Rui, Zhang

doi:10.1177/1460458219867231

Cited by 15 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Interestingly, consuming micronutrients post-disaster including earthquakes, floods and massacres showed that micronutrient supplements reduced the risk of PTSD from 75 to 15% [ 53 ]. However, one study using Twitter Sentiment Analysis concluded an individuals with mental health disorders and taking a supplement showed more negative emotions in their tweets [ 54 ]. Unfortunately, this study did not measure the level of micronutrients consumed by the respondents; thus, the cause and effect between hypervigilance and consuming dietary supplements could not be concluded.…”

Section: Discussionmentioning

confidence: 99%

Societal Influence and Psychological Distress Among Indonesian Adults in Java on the Early Omicron Wave of COVID-19

d'Arqom,

Akram,

Azzahranisa

et al. 2023

Future Sci. OA

View full text Add to dashboard Cite

Aim: Changes during the COVID-19 pandemic might create pressure on different people, thus this study aimed to measure respondents' psychological distress during the early phase of the Omicron wave in Java Island. Methods: A web-based survey on societal influence and COVID-19-related psychological distress was distributed through social media in November–December 2021, and received 396 responses. Results: This study showed that almost 50% of respondents faced psychological distress during the early phase of the Omicron variant, especially concerning hypervigilance and avoidance. Several sociodemographic factors might contribute to the incidence of psychological distress including comorbidity, age and education. Conclusion: Taken together, the incidence of COVID-19-related psychological distress was still found in the early phase of the Omicron variant, especially among young adults.

show abstract

Section: Discussionmentioning

confidence: 99%

Societal Influence and Psychological Distress Among Indonesian Adults in Java on the Early Omicron Wave of COVID-19

d'Arqom,

Akram,

Azzahranisa

et al. 2023

Future Sci. OA

View full text Add to dashboard Cite

show abstract

“…The two most notable linguistic dimensions related to sentiment are “positive emotion” and “negative emotion”, both of which have been used in several studies to measure positive and negative emotion in the online users’ posts [ 40 , 41 ]. In this paper, sentiment polarity included three types: positive, neutral, and negative.…”

Section: Methodsmentioning

confidence: 99%

Do Informational and Emotional Elements Differ between Online Psychological and Physiological Disease Communities in China? A Comparative Study of Depression and Diabetes

Yao

Zhang

et al. 2022

IJERPH

View full text Add to dashboard Cite

Disease-specific online health communities provide a convenient and common platform for patients to share experiences, change information, provide and receive social support. This study aimed to compare differences between online psychological and physiological disease communities in topics, sentiment, participation, and emotional contagion patterns using multiple methods as well as to discuss how to satisfy the users’ different informational and emotional needs. We chose the online depression and diabetes communities on the Baidu Tieba platform as the data source. Topic modeling and theme coding were employed to analyze discussion preferences for various topic categories. Sentiment analysis was used to identify the sentiment polarity of each post and comment. The social network was used to represent the users’ interaction and emotional flows to discover the differences in participation and emotional contagion patterns between psychological and physiological disease communities. The results revealed that people affected by depression focused more on their symptoms and social relationships, while people affected by diabetes were more likely to discuss treatment and self-management behavior. In the depression community, there were obvious interveners spreading positive emotions and more core users in the negative emotional contagion network. In the diabetes community, emotional contagion was less prevalent and core users in positive and negative emotional contagion networks were basically the same. The study reveals insights into the differences between online psychological and physiological disease communities, providing a greater understanding of the users’ informational and emotional needs expressed online. These results are helpful for society to provide actual medical assistance and deploy health interventions based on disease types.

show abstract

“…We selected only English tweets to develop the concept extraction and relation extraction models. A total of 247,807 tweets that satisfy the criteria were found from a Twitter database we constructed in prior work [34,35] using the Twitter streaming application programming interface (API), covering daily publictweets from 2012 to 2018.…”

Section: Data Collectionmentioning

confidence: 99%

Deep learning models in detection of dietary supplement adverse event signals from Twitter

et al. 2021

Self Cite

View full text Add to dashboard Cite

Objective The objective of this study is to develop a deep learning pipeline to detect signals on dietary supplement-related adverse events (DS AEs) from Twitter. Materials and Methods We obtained 247 807 tweets ranging from 2012 to 2018 that mentioned both DS and AE. We designed a tailor-made annotation guideline for DS AEs and annotated biomedical entities and relations on 2000 tweets. For the concept extraction task, we fine-tuned and compared the performance of BioClinical-BERT, PubMedBERT, ELECTRA, RoBERTa, and DeBERTa models with a CRF classifier. For the relation extraction task, we fine-tuned and compared BERT models to BioClinical-BERT, PubMedBERT, RoBERTa, and DeBERTa models. We chose the best-performing models in each task to assemble an end-to-end deep learning pipeline to detect DS AE signals and compared the results to the known DS AEs from a DS knowledge base (ie, iDISK). Results DeBERTa-CRF model outperformed other models in the concept extraction task, scoring a lenient microaveraged F1 score of 0.866. RoBERTa model outperformed other models in the relation extraction task, scoring a lenient microaveraged F1 score of 0.788. The end-to-end pipeline built on these 2 models was able to extract DS indication and DS AEs with a lenient microaveraged F1 score of 0.666. Conclusion We have developed a deep learning pipeline that can detect DS AE signals from Twitter. We have found DS AEs that were not recorded in an existing knowledge base (iDISK) and our proposed pipeline can as sist DS AE pharmacovigilance.

show abstract

Detecting associations between dietary supplement intake and sentiments within mental disorder tweets

Cited by 15 publications

References 12 publications

Societal Influence and Psychological Distress Among Indonesian Adults in Java on the Early Omicron Wave of COVID-19

Societal Influence and Psychological Distress Among Indonesian Adults in Java on the Early Omicron Wave of COVID-19

Do Informational and Emotional Elements Differ between Online Psychological and Physiological Disease Communities in China? A Comparative Study of Depression and Diabetes

Deep learning models in detection of dietary supplement adverse event signals from Twitter

Contact Info

Product

Resources

About