Dataset Information

Comparing Emotional Valence Scores of Twitter Messages from Human Coding and Machine Learning Algorithms Among Hispanic and African American Family Caregivers of Persons with Dementia.

ABSTRACT: We compared emotional valence scores as determined via machine learning approaches to human-coded scores of direct messages on Twitter from our 2,301 followers during a Twitter-based clinical trial screening for Hispanic and African American family caregivers of persons with dementia. We manually assigned emotional valence scores to 249 randomly selected direct Twitter messages from our followers (N=2,301), then we applied three machine learning sentiment analysis algorithms to extract emotional valence scores for each message and compared their mean scores to the human coding results. The aggregated mean emotional scores from the natural language processing were slightly positive, while the mean score from human coding as a gold standard was negative. Clusters of strongly negative sentiments were observed in followers' responses to being found non-eligible for the study, indicating a significant need for alternative strategies to provide similar research opportunities to non-eligible family caregivers.

SUBMITTER: Yoon S

PROVIDER: S-EPMC10363991 | biostudies-literature | 2023 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Comparing Emotional Valence Scores of Twitter Messages from Human Coding and Machine Learning Algorithms Among Hispanic and African American Family Caregivers of Persons with Dementia.

Yoon Sunmoo S Broadwell Peter P Tipiani Dante D Bristol Amanda A Moon Soyoung S Yoon Brian B Liu Jianfang J Huang Niya N Davis Nicole N

Studies in health technology and informatics 20230601

We compared emotional valence scores as determined via machine learning approaches to human-coded scores of direct messages on Twitter from our 2,301 followers during a Twitter-based clinical trial screening for Hispanic and African American family caregivers of persons with dementia. We manually assigned emotional valence scores to 249 randomly selected direct Twitter messages from our followers (N=2,301), then we applied three machine learning sentiment analysis algorithms to extract emotional ...[more]

PMID: 37387060

Similar Datasets

Project description:IntroductionThe prevalence of dementia is increasing without a known cure, resulting in an increasing number of informal caregivers. Caring for a person with dementia results in increased stress and depressive symptoms. There are several behavioural interventions designed to alleviate stress and depressive symptoms in caregivers of persons with dementia with evidence of efficacy. Two of the best-known interventions are the New York University Caregiver Intervention (NYUCI) and the Resources for Enhancing Alzheimer's Caregivers Health (REACH). The effectiveness of the NYUCI and REACH has never been compared. There is also a paucity of data on which interventions are more effective in Hispanics in New York City. Thus, we proposed the Northern Manhattan Hispanic Caregiver intervention Effectiveness Study (NHiCE), a pragmatic clinical trial designed to compare the effectiveness of adaptations of the NYUCI and the REACH in informal Hispanic caregivers of persons with dementia in New York City.Methods and analysisNHiCE is a 6-month randomised controlled trial comparing the effectiveness of adaptations of the NYUCI and REACH among 200 Hispanic informal adult caregivers of persons with dementia. The planned number of sessions of the NYUCI and REACH are similar. The primary outcome measures are changes from baseline to 6 months in the Zarit Caregiver Burden Scale and Geriatric Depression Scale. Our primary approach to analyses will be intent-to-treat. The primary analyses will use mixed random effects models, and a full information maximum likelihood approach, with sensitivity analyses using generalised estimating equation.Ethics and disseminationNHiCE is approved by the Institutional Review Board of Columbia University Medical Center (protocol AAAM5150). A Data Safety Monitoring Board monitors the progress of the study. Dissemination will include reports of the characteristics of the study participants, as well as a report of the results of the clinical trial.Trial registration numberNCT02092987, Pre-results.

Project description:BackgroundThe Older Persons and Informal Caregivers Survey-Minimum Dataset (TOPICS-MDS) collects uniform information from research projects funded under the Dutch National Care for the Elderly Programme. To compare the effectiveness of these projects a preference-weighted outcome measure that combined multidimensional TOPICS-MDS outcomes into a composite endpoint (TOPICS-CEP) was developed based on the health state preferences of older persons and informal caregivers.ObjectivesTo derive preference weights for TOPICS-CEP's components based on health state preferences of healthcare professionals and to investigate whether these weights differ between disciplines and differ from those of older persons and informal caregivers.Materials and methodsVignette studies were conducted. Participants assessed the general wellbeing of older persons described in vignettes on a scale (0-10). Mixed linear analyses were used to obtain and compare the preference weights of the eight TOPICS-CEP components: morbidities, functional limitations, emotional wellbeing, pain experience, cognitive problems, social functioning, self-perceived health, and self-perceived quality of life (QOL).ResultsOverall, 330 healthcare professionals, 124 older persons and 76 informal caregivers participated. The preference weights were not significantly different between disciplines. However, the professionals' preference weights differed significantly from those of older persons and informal caregivers. Morbidities and functional limitations were given more weight by older persons and informal caregivers than by healthcare professionals [difference between preference weights: 0.12 and 0.07] while the opposite was true for pain experience, social functioning, and self-perceived QOL [difference between preference weights: 0.13, 0.15 and 0.26].ConclusionIt is important to recognize the discrepancies between the health state preferences of various stakeholders to (1) correctly interpret results when studying the effectiveness of interventions in elderly care and (2) establish appropriate healthcare policies. Furthermore, we should strive to include older persons in our decision making process through a shared decision making approach.

Dataset Information

Comparing Emotional Valence Scores of Twitter Messages from Human Coding and Machine Learning Algorithms Among Hispanic and African American Family Caregivers of Persons with Dementia.

Publications

Comparing Emotional Valence Scores of Twitter Messages from Human Coding and Machine Learning Algorithms Among Hispanic and African American Family Caregivers of Persons with Dementia.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets