FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4213306621> ?p ?o ?g. }

• W4213306621 abstract "<sec> <title>BACKGROUND</title> Mental illness affects a significant portion of the worldwide population. Online mental health forums can provide a supportive environment for those afflicted and also generate a large amount of data that can be mined to predict mental health states using machine learning methods. </sec> <sec> <title>OBJECTIVE</title> This study aimed to benchmark multiple methods of text feature representation for social media posts and compare their downstream use with automated machine learning (AutoML) tools. We tested on datasets that contain posts labeled for perceived suicide risk or moderator attention in the context of self-harm. Specifically, we assessed the ability of the methods to prioritize posts that a moderator would identify for immediate response. </sec> <sec> <title>METHODS</title> We used 1588 labeled posts from the Computational Linguistics and Clinical Psychology (CLPsych) 2017 shared task collected from the Reachout.com forum. Posts were represented using lexicon-based tools, including Valence Aware Dictionary and sEntiment Reasoner, Empath, and Linguistic Inquiry and Word Count, and also using pretrained artificial neural network models, including DeepMoji, Universal Sentence Encoder, and Generative Pretrained Transformer-1 (GPT-1). We used Tree-based Optimization Tool and Auto-Sklearn as AutoML tools to generate classifiers to triage the posts. </sec> <sec> <title>RESULTS</title> The top-performing system used features derived from the GPT-1 model, which was fine-tuned on over 150,000 unlabeled posts from Reachout.com. Our top system had a macroaveraged F1 score of 0.572, providing a new state-of-the-art result on the CLPsych 2017 task. This was achieved without additional information from metadata or preceding posts. Error analyses revealed that this top system often misses expressions of hopelessness. In addition, we have presented visualizations that aid in the understanding of the learned classifiers. </sec> <sec> <title>CONCLUSIONS</title> In this study, we found that transfer learning is an effective strategy for predicting risk with relatively little labeled data and noted that fine-tuning of pretrained language models provides further gains when large amounts of unlabeled text are available. </sec>" @default.
• W4213306621 created "2022-02-24" @default.
• W4213306621 creator A5025071606 @default.
• W4213306621 creator A5041523640 @default.
• W4213306621 creator A5046916534 @default.
• W4213306621 creator A5054607778 @default.
• W4213306621 creator A5066503034 @default.
• W4213306621 creator A5074657982 @default.
• W4213306621 date "2019-07-04" @default.
• W4213306621 modified "2023-10-16" @default.
• W4213306621 title "Transfer Learning for Risk Classification of Social Media Posts: Model Evaluation Study (Preprint)" @default.
• W4213306621 cites W2097609979 @default.
• W4213306621 cites W2129605504 @default.
• W4213306621 cites W2140910804 @default.
• W4213306621 cites W2162051395 @default.
• W4213306621 cites W2273847690 @default.
• W4213306621 cites W2338065342 @default.
• W4213306621 cites W2554980225 @default.
• W4213306621 cites W2593668264 @default.
• W4213306621 cites W2611552941 @default.
• W4213306621 cites W2807452501 @default.
• W4213306621 cites W2884435597 @default.
• W4213306621 cites W2898410082 @default.
• W4213306621 doi "https://doi.org/10.2196/preprints.15371" @default.
• W4213306621 hasPublicationYear "2019" @default.
• W4213306621 type Work @default.
• W4213306621 citedByCount "0" @default.
• W4213306621 crossrefType "posted-content" @default.
• W4213306621 hasAuthorship W4213306621A5025071606 @default.
• W4213306621 hasAuthorship W4213306621A5041523640 @default.
• W4213306621 hasAuthorship W4213306621A5046916534 @default.
• W4213306621 hasAuthorship W4213306621A5054607778 @default.
• W4213306621 hasAuthorship W4213306621A5066503034 @default.
• W4213306621 hasAuthorship W4213306621A5074657982 @default.
• W4213306621 hasBestOaLocation W42133066212 @default.
• W4213306621 hasConcept C119857082 @default.
• W4213306621 hasConcept C136764020 @default.
• W4213306621 hasConcept C148524875 @default.
• W4213306621 hasConcept C154945302 @default.
• W4213306621 hasConcept C204321447 @default.
• W4213306621 hasConcept C23123220 @default.
• W4213306621 hasConcept C2777462759 @default.
• W4213306621 hasConcept C41008148 @default.
• W4213306621 hasConcept C41608201 @default.
• W4213306621 hasConcept C518677369 @default.
• W4213306621 hasConceptScore W4213306621C119857082 @default.
• W4213306621 hasConceptScore W4213306621C136764020 @default.
• W4213306621 hasConceptScore W4213306621C148524875 @default.
• W4213306621 hasConceptScore W4213306621C154945302 @default.
• W4213306621 hasConceptScore W4213306621C204321447 @default.
• W4213306621 hasConceptScore W4213306621C23123220 @default.
• W4213306621 hasConceptScore W4213306621C2777462759 @default.
• W4213306621 hasConceptScore W4213306621C41008148 @default.
• W4213306621 hasConceptScore W4213306621C41608201 @default.
• W4213306621 hasConceptScore W4213306621C518677369 @default.
• W4213306621 hasLocation W42133066211 @default.
• W4213306621 hasLocation W42133066212 @default.
• W4213306621 hasOpenAccess W4213306621 @default.
• W4213306621 hasPrimaryLocation W42133066211 @default.
• W4213306621 hasRelatedWork W2398825887 @default.
• W4213306621 hasRelatedWork W2618302280 @default.
• W4213306621 hasRelatedWork W2901590103 @default.
• W4213306621 hasRelatedWork W2947903144 @default.
• W4213306621 hasRelatedWork W2984206076 @default.
• W4213306621 hasRelatedWork W3107474891 @default.
• W4213306621 hasRelatedWork W3194539120 @default.
• W4213306621 hasRelatedWork W4312833533 @default.
• W4213306621 hasRelatedWork W4321496520 @default.
• W4213306621 hasRelatedWork W4323060038 @default.
• W4213306621 isParatext "false" @default.
• W4213306621 isRetracted "false" @default.
• W4213306621 workType "article" @default.