Sentiment Analysis of Insomnia-Related Tweets via a Combination of Transformers Using Dempster-Shafer Theory : Pre- and Peri-COVID-19 Pandemic Retrospective Study
©Arash Maghsoudi, Sara Nowakowski, Ritwick Agrawal, Amir Sharafkhaneh, Mark E Kunik, Aanand D Naik, Hua Xu, Javad Razjouyan. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 27.12.2022..
BACKGROUND: The COVID-19 pandemic has imposed additional stress on population health that may result in a change of sleeping behavior.
OBJECTIVE: In this study, we hypothesized that using natural language processing to explore social media would help with assessing the mental health conditions of people experiencing insomnia after the outbreak of COVID-19.
METHODS: We designed a retrospective study that used public social media content from Twitter. We categorized insomnia-related tweets based on time, using the following two intervals: the prepandemic (January 1, 2019, to January 1, 2020) and peripandemic (January 1, 2020, to January 1, 2021) intervals. We performed a sentiment analysis by using pretrained transformers in conjunction with Dempster-Shafer theory (DST) to classify the polarity of emotions as positive, negative, and neutral. We validated the proposed pipeline on 300 annotated tweets. Additionally, we performed a temporal analysis to examine the effect of time on Twitter users' insomnia experiences, using logistic regression.
RESULTS: We extracted 305,321 tweets containing the word insomnia (prepandemic tweets: n=139,561; peripandemic tweets: n=165,760). The best combination of pretrained transformers (combined via DST) yielded 84% accuracy. By using this pipeline, we found that the odds of posting negative tweets (odds ratio [OR] 1.39, 95% CI 1.37-1.41; P<.001) were higher in the peripandemic interval compared to those in the prepandemic interval. The likelihood of posting negative tweets after midnight was 21% higher than that before midnight (OR 1.21, 95% CI 1.19-1.23; P<.001). In the prepandemic interval, while the odds of posting negative tweets were 2% higher after midnight compared to those before midnight (OR 1.02, 95% CI 1.00-1.07; P=.008), they were 43% higher (OR 1.43, 95% CI 1.40-1.46; P<.001) in the peripandemic interval.
CONCLUSIONS: The proposed novel sentiment analysis pipeline, which combines pretrained transformers via DST, is capable of classifying the emotions and sentiments of insomnia-related tweets. Twitter users shared more negative tweets about insomnia in the peripandemic interval than in the prepandemic interval. Future studies using a natural language processing framework could assess tweets about other types of psychological distress, habit changes, weight gain resulting from inactivity, and the effect of viral infection on sleep.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2022 |
|---|---|
| Erschienen: |
2022 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:24 |
|---|---|
| Enthalten in: |
Journal of medical Internet research - 24(2022), 12 vom: 27. Dez., Seite e41517 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Maghsoudi, Arash [VerfasserIn] |
|---|
| Links: |
|---|
| Anmerkungen: |
Date Completed 29.12.2022 Date Revised 13.05.2023 published: Electronic Citation Status MEDLINE |
|---|
| doi: |
10.2196/41517 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM349306850 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM349306850 | ||
| 003 | DE-627 | ||
| 005 | 20231226042315.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2022 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.2196/41517 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1164.xml |
| 035 | |a (DE-627)NLM349306850 | ||
| 035 | |a (NLM)36417585 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Maghsoudi, Arash |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Sentiment Analysis of Insomnia-Related Tweets via a Combination of Transformers Using Dempster-Shafer Theory |b Pre- and Peri-COVID-19 Pandemic Retrospective Study |
| 264 | 1 | |c 2022 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 29.12.2022 | ||
| 500 | |a Date Revised 13.05.2023 | ||
| 500 | |a published: Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a ©Arash Maghsoudi, Sara Nowakowski, Ritwick Agrawal, Amir Sharafkhaneh, Mark E Kunik, Aanand D Naik, Hua Xu, Javad Razjouyan. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 27.12.2022. | ||
| 520 | |a BACKGROUND: The COVID-19 pandemic has imposed additional stress on population health that may result in a change of sleeping behavior | ||
| 520 | |a OBJECTIVE: In this study, we hypothesized that using natural language processing to explore social media would help with assessing the mental health conditions of people experiencing insomnia after the outbreak of COVID-19 | ||
| 520 | |a METHODS: We designed a retrospective study that used public social media content from Twitter. We categorized insomnia-related tweets based on time, using the following two intervals: the prepandemic (January 1, 2019, to January 1, 2020) and peripandemic (January 1, 2020, to January 1, 2021) intervals. We performed a sentiment analysis by using pretrained transformers in conjunction with Dempster-Shafer theory (DST) to classify the polarity of emotions as positive, negative, and neutral. We validated the proposed pipeline on 300 annotated tweets. Additionally, we performed a temporal analysis to examine the effect of time on Twitter users' insomnia experiences, using logistic regression | ||
| 520 | |a RESULTS: We extracted 305,321 tweets containing the word insomnia (prepandemic tweets: n=139,561; peripandemic tweets: n=165,760). The best combination of pretrained transformers (combined via DST) yielded 84% accuracy. By using this pipeline, we found that the odds of posting negative tweets (odds ratio [OR] 1.39, 95% CI 1.37-1.41; P<.001) were higher in the peripandemic interval compared to those in the prepandemic interval. The likelihood of posting negative tweets after midnight was 21% higher than that before midnight (OR 1.21, 95% CI 1.19-1.23; P<.001). In the prepandemic interval, while the odds of posting negative tweets were 2% higher after midnight compared to those before midnight (OR 1.02, 95% CI 1.00-1.07; P=.008), they were 43% higher (OR 1.43, 95% CI 1.40-1.46; P<.001) in the peripandemic interval | ||
| 520 | |a CONCLUSIONS: The proposed novel sentiment analysis pipeline, which combines pretrained transformers via DST, is capable of classifying the emotions and sentiments of insomnia-related tweets. Twitter users shared more negative tweets about insomnia in the peripandemic interval than in the prepandemic interval. Future studies using a natural language processing framework could assess tweets about other types of psychological distress, habit changes, weight gain resulting from inactivity, and the effect of viral infection on sleep | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Research Support, N.I.H., Extramural | |
| 650 | 4 | |a Research Support, U.S. Gov't, Non-P.H.S. | |
| 650 | 4 | |a COVID-19 | |
| 650 | 4 | |a Dempster-Shafer theory | |
| 650 | 4 | |a Twitter | |
| 650 | 4 | |a coronavirus | |
| 650 | 4 | |a effect | |
| 650 | 4 | |a natural language processing | |
| 650 | 4 | |a pandemic | |
| 650 | 4 | |a sentiment analysis | |
| 650 | 4 | |a sleep | |
| 650 | 4 | |a sleeping | |
| 650 | 4 | |a social media | |
| 650 | 4 | |a transformers | |
| 650 | 4 | |a viral infection | |
| 700 | 1 | |a Nowakowski, Sara |e verfasserin |4 aut | |
| 700 | 1 | |a Agrawal, Ritwick |e verfasserin |4 aut | |
| 700 | 1 | |a Sharafkhaneh, Amir |e verfasserin |4 aut | |
| 700 | 1 | |a Kunik, Mark E |e verfasserin |4 aut | |
| 700 | 1 | |a Naik, Aanand D |e verfasserin |4 aut | |
| 700 | 1 | |a Xu, Hua |e verfasserin |4 aut | |
| 700 | 1 | |a Razjouyan, Javad |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Journal of medical Internet research |d 1999 |g 24(2022), 12 vom: 27. Dez., Seite e41517 |w (DE-627)NLM116127104 |x 1438-8871 |7 nnns |
| 773 | 1 | 8 | |g volume:24 |g year:2022 |g number:12 |g day:27 |g month:12 |g pages:e41517 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.2196/41517 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 24 |j 2022 |e 12 |b 27 |c 12 |h e41517 | ||