A Friend in Need Is a Friend Indeed: Investigating the Quality of Training Data from Peers for Auto-generating Empathetic Textual Responses to Non-Sensitive Posts in a Cohort of College Students DOI Open Access
Ravi Sharma, Jamshidbek Mirzakhalov, Pratool Bharti

et al.

ACM Journal on Computing and Sustainable Societies, Journal Year: 2023, Volume and Issue: 1(2), P. 1 - 27

Published: Sept. 16, 2023

Towards providing personalized care, digital mental-wellness apps today ask questions to learn about subjects. However, not all subjects using these will have mood problems; thus, they do need follow-up questions. In this study, we investigate an alternate mechanism handle such non-sensitive posts (i.e., those indicating problems) in college settings. To so, generate and use training data provided by a cohort of peer students so that responses are contextual, emotionally aware, empathetic while also being terminal (not asking questions). Using from real app used students, identify AI models trained with our peer-provided dataset desirable posts, state-of-the-art (Facebook’s) Empathetic Dataset yields many questions, hence giving perception intrusive. We believe mental wellness must assume any subject has problems. Perceptions intrusiveness questions) be factor design. can provide rich reliable datasets for apps, topic is yet explored.

Language: Английский

This and That in Depression: Cross-linguistic Semantic Effects DOI Open Access
Line Kruse, Roberta Rocca, Emanuela Todisco

et al.

Published: Aug. 17, 2024

Demonstratives (in English "this" and "that") are pivotal to human communication, facilitating joint attention the establishment of a common ground reference. All languages have at least two forms, typically distinguishing proximal from distal space, where "space" is defined by range context-dependent physical, psychological, social referent-intrinsic factors. Recent work based on Demonstrative Choice Task (DCT) has indicated, that in absence guiding context, demonstrative reference may capture an experienced or emotional proximity referent concepts, semantic differences responses allow implicit inferences individual psychological space related depression. The present paper investigated extent which these patterns generalize across languages, including German, Spanish, Italian, Russian, Chinese Tagalog Filipino samples. DCT-based classification models robustly outperformed baseline all except for Chinese, showed similar as observed English. Particularly negative emotion features were consistently among most important models, choice form was more frequent depression group than control group. opposite pattern positive features, however, effects variable languages. Results suggest simple lexical choices DCT experiential states depression, be used map individuals along broad potentially providing novel insights into disorder etiology.

Language: Английский

Citations

0
Sentiment Analysis of Tweets Towards Vaccines via Bidirectional Encoder Representations from Transformers (BERT) DOI
Joel C. De Goma,

Rosselle Mae Mercado,

Timothy Jacob Saporna

et al.

Published: April 26, 2024

Language: Английский

Citations

0
Examining the Effects of Static Personality Traits with Dynamic Affective and Emotional States on Depression Severity DOI
Abdullah Ahmed, Jayroop Ramesh, Sandipan Ganguly

et al.

Lecture notes in networks and systems, Journal Year: 2024, Volume and Issue: unknown, P. 491 - 497

Published: Oct. 22, 2024

Language: Английский

Citations

0
Unifying Perspectives: CNN-LSTM Integration for Anxiety and Depression Prediction Through Textual Analysis DOI

Sharon Susan Jacob

Lecture notes in networks and systems, Journal Year: 2024, Volume and Issue: unknown, P. 219 - 232

Published: Jan. 1, 2024

Language: Английский

Citations

0
A Friend in Need Is a Friend Indeed: Investigating the Quality of Training Data from Peers for Auto-generating Empathetic Textual Responses to Non-Sensitive Posts in a Cohort of College Students DOI Open Access
Ravi Sharma, Jamshidbek Mirzakhalov, Pratool Bharti

et al.

ACM Journal on Computing and Sustainable Societies, Journal Year: 2023, Volume and Issue: 1(2), P. 1 - 27

Published: Sept. 16, 2023

Towards providing personalized care, digital mental-wellness apps today ask questions to learn about subjects. However, not all subjects using these will have mood problems; thus, they do need follow-up questions. In this study, we investigate an alternate mechanism handle such non-sensitive posts (i.e., those indicating problems) in college settings. To so, generate and use training data provided by a cohort of peer students so that responses are contextual, emotionally aware, empathetic while also being terminal (not asking questions). Using from real app used students, identify AI models trained with our peer-provided dataset desirable posts, state-of-the-art (Facebook’s) Empathetic Dataset yields many questions, hence giving perception intrusive. We believe mental wellness must assume any subject has problems. Perceptions intrusiveness questions) be factor design. can provide rich reliable datasets for apps, topic is yet explored.

Language: Английский

Citations

0