Background : An unproven “nicotine hypothesis” that indicates nicotine's therapeutic potential for COVID-19 has been proposed in recent literature. This study is about Twitter posts that misinterpret this hypothesis to make baseless claims about benefits of smoking and vaping in the context of COVID-19. We quantify the presence of such misinformation and characterize the tweeters who post such messages. Methods : Twitter premium API was used to download tweets (n = 17,533) that match terms indicating (a) nicotine or vaping themes, (b) a prophylactic or therapeutic effect, and (c) COVID-19 (January-July 2020) as a conjunctive query. A constraint on the length of the span of text containing the terms in the tweets allowed us to focus on those that convey the therapeutic intent. We hand-annotated these filtered tweets and built a classifier that identifies tweets that extrapolate the nicotine hypothesis to smoking/vaping with a positive predictive value of 85%. We analyzed the frequently used terms in author bios, top Web links, and hashtags of such tweets. Results : 21% of our filtered COVID-19 tweets indicate a vaping or smoking-based prevention/treatment narrative. Qualitative analyses show a variety of ways therapeutic claims are being made and tweeter bios reveal pre-existing notions of positive stances toward vaping. Conclusion : The social media landscape is a double-edged sword in tobacco communication. Although it increases information reach, consumers can also be subject to confirmation bias when exposed to inadvertent or deliberate framing of scientific discourse that may border on misinformation. This calls for circumspection and additional planning in countering such narratives as the COVID-19 pandemic continues to ravage our world. Our results also serve as a cautionary tale in how social media can be leveraged to spread misleading information about tobacco products in the wake of pandemics.
ObjectiveThe low observed prevalence of smokers among hospitalized COVID-19 patients in certain cohorts has led to a hypothesis regarding nicotine’s therapeutic role in COVID-19 prevention and treatment. As new scientific evidence surfaces, premature conclusions about nicotine are rife in social media, especially unwarranted leaps of such associations to vaping and smoking. This study reports on the prevalence of such leaps and the nature of authors who are making them.MethodsWe used a Twitter API subscription service to download tweets (n = 17,533) that match terms indicating nicotine or vaping themes, in addition to those that point to a prophylactic or therapeutic effect and COVID-19 (January-July 2020). Using a windowing approach, we focused on tweets that are more likely to convey the therapeutic intent. We hand-annotated these filtered tweets and built a classifier that identifies tweets that extrapolate a nicotine link to vaping/smoking. We analyzed the frequently used terms in author bios, top Web links, and hashtags of such tweets.Results21% of our filtered tweets indicate a vaping/smoking-based prevention/treatment narrative. Our classifier was able to spot tweets that make unproven claims about vaping/smoking and COVID-19 with a positive predictive value of 85%. Qualitative analyses show a variety of ways therapeutic claims are being made and user bios reveal pre-existing notions of positive stances toward vaping.ConclusionThe social media landscape is a double-edged sword in tobacco communication. Although it increases information reach, consumers can also be subject to confirmation bias when exposed to inadvertent or deliberate framing of scientific discourse that may border on misinformation. This calls for circumspection and additional planning in countering such narratives as the COVID-19 pandemic continues to ravage our world.
Information retrieval (IR) for precision medicine (PM) often involves looking for multiple pieces of evidence that characterize a patient case. This typically includes at least the name of a condition and a genetic variation that applies to the patient. Other factors such as demographic attributes, comorbidities, and social determinants may also be pertinent. As such, the retrieval problem is often formulated as ad hoc search but with multiple facets (e.g., disease, mutation) that may need to be incorporated. In this paper, we present a document reranking approach that combines neural query-document matching and text summarization toward such retrieval scenarios. Our architecture builds on the basic BERT model with three specific components for reranking: (a). document-query matching (b). keyword extraction and (c). facet-conditioned abstractive summarization. The outcomes of (b) and (c) are used to essentially transform a candidate document into a concise summary that can be compared with the query at hand to compute a relevance score. Component (a) directly generates a matching score of a candidate document for a query. The full architecture benefits from the complementary potential of document-query matching and the novel document transformation approach based on summarization along PM facets. Evaluations using NIST's TREC-PM track datasets (2017-2019) show that our model achieves state-of-the-art performance. To foster reproducibility, our code is made available here: https://github.com/bionlproc/t ext-summ-for-doc-retrieval.
Document retrieval (DR) forms an important component in end-to-end question-answering (QA) systems where particular answers are sought for well-formed questions. DR in the QA scenario is also useful by itself even without a more involved natural language processing component to extract exact answers from the retrieved documents. This latter step may simply be done by humans like in traditional search engines granted the retrieved documents contain the answer. In this paper, we take advantage of datasets made available through the BioASQ end-to-end QA shared task series and build an effective biomedical DR system that relies on relevant answer snippets in the BioASQ training datasets. At the core of our approach is a question-answer sentence matching neural network that learns a measure of relevance of a sentence to an input question in the form of a matching score. In addition to this matching score feature, we also exploit two auxiliary features for scoring document relevance: the name of the journal in which a document is published and the presence/absence of semantic relations (subject-predicate-object triples) in a candidate answer sentence connecting entities mentioned in the question. We rerank our baseline sequential dependence model scores using these three additional features weighted via adaptive random research and other learning-to-rank methods. Our full system placed 2nd in the final batch of Phase A (DR) of task B (QA) in BioASQ 2018. Our ablation experiments highlight the significance of the neural matching network component in the full system.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.