Exploring Self-Identified Counseling Expertise in Online Support Forums

Lahnala, Allison; Zhao, Yuntian; Welch, Charles; Kummerfeld, Jonathan K.; An, Lawrence C.; Resnicow, Kenneth; Mihalcea, Rada; Pérez-Rosas, Verónica

doi:10.18653/v1/2021.findings-acl.392

Cited by 11 publications

(9 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, we believe it would be better to use a toxicity dataset that includes conversational context. Our improvements to mitigation of toxic degeneration could be better understood and further expanded upon in a conversational application where empathy is important, such as counseling or online mental health support (Sharma et al, 2021;Lahnala et al, 2021).…”

Section: Discussion and Limitationsmentioning

confidence: 99%

Mitigating Toxic Degeneration with Empathetic Data: Exploring the Relationship Between Toxicity and Empathy

Lahnala¹,

Welch²,

Neuendorf³

et al. 2022

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

Self Cite

View full text Add to dashboard Cite

Content Warning: This paper includes examples of religious-based discriminatory language that may be offensive and upsetting.Large pre-trained neural language models have supported the effectiveness of many NLP tasks, yet are still prone to generating toxic language hindering the safety of their use. Using empathetic data, we improve over recent work on controllable text generation that aims to reduce the toxicity of generated text. We find we are able to dramatically reduce the size of finetuning data to 7.5-30k samples while at the same time making significant improvements over state-of-the-art toxicity mitigation of up to 3.4% absolute reduction (26% relative) from the original work on 2.3m samples, by strategically sampling data based on empathy scores. We observe that the degree of improvement is subject to specific communication components of empathy. In particular, the cognitive components of empathy significantly beat the original dataset in almost all experiments, while emotional empathy was tied to less improvement and even underperforming random samples of the original data. This is a particularly implicative insight for NLP work concerning empathy as until recently the research and resources built for it have exclusively considered empathy as an emotional concept.

show abstract

Section: Discussion and Limitationsmentioning

confidence: 99%

Mitigating Toxic Degeneration with Empathetic Data: Exploring the Relationship Between Toxicity and Empathy

Lahnala¹,

Welch²,

Neuendorf³

et al. 2022

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our work differs in the sense that our end-to-end system does not provide any clinical suggestions or therapy recommendations. The role of competence in responses to help-seeking posts on mental health was investigated in (Lahnala et al, 2021).…”

Section: Related Workmentioning

confidence: 99%

“…Some of the open-sourced datasets such as DAIC-WOZ (Gratch et al, 2014) contained small-scale conversations. Recent support-based datasets (Sharma et al, 2020;Lahnala et al, 2021), on the other hand, featured pairs of seeker post and supporter response with no dialogic structure. Inspired by previous research, we create the Mo-tiVAte dataset, which was acquired via a peer-topeer support platform and is ideal for our objective.…”

Section: Related Workmentioning

confidence: 99%

A Shoulder to Cry on: Towards A Motivational Virtual Assistant for Assuaging Mental Agony

Saha¹,

Reddy²,

Das³

et al. 2022

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

View full text Add to dashboard Cite

Mental Health Disorders continue plaguing humans worldwide. Aggravating this situation is the severe shortage of qualified and competent mental health professionals (MHPs), which underlines the need for developing Virtual Assistants (VAs) that can assist MHPs. The data+ML for automation can come from platforms that allow visiting and posting messages in peerto-peer anonymous manner for sharing their experiences (frequently stigmatized) and seeking support. In this paper, we propose a VA that can act as the first point of contact and comfort for mental health patients. We curate a dataset, Motivational VA: MotiVAte comprising of 7k dyadic conversations collected from a peer-topeer support platform. The system employs two mechanisms: (i) Mental Illness Classification: an attention based BERT classifier that outputs the mental disorder category out of the 4 categories, viz., Major Depressive Disorder (MDD), Anxiety, Obsessive Compulsive Disorder (OCD) and Post-traumatic Stress Disorder (PTSD), based on the input ongoing dialog between the support seeker and the VA; and (ii) Mental Illness Conditioned Motivational Dialogue Generation (MI-MDG): a sentiment driven Reinforcement Learning (RL) based motivational response generator. The empirical evaluation demonstrates the system capability by way of outperforming several baselines.

show abstract

“…This dataset, originally collected from so-called "Subreddits" of the social media platform Reddit, consists of posts and comments authored in environments discussing mental health issues (Lahnala et al, 2021). In order to ensure enough content to properly model each user, we further filter the dataset to only include users with at least 100 posts and/or comments, leaving us with a total of 5,299 users.…”

Section: Mhp Datasetmentioning

confidence: 99%

Modeling Proficiency with Implicit User Representations

Breitwieser,

Lahnala,

Welch

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

We introduce the problem of proficiency modeling: Given a user's posts on a social media platform, the task is to identify the subset of posts or topics for which the user has some level of proficiency. This enables the filtering and ranking of social media posts on a given topic as per user proficiency. Unlike experts on a given topic, proficient users may not have received formal training and possess years of practical experience, but may be autodidacts, hobbyists, and people with sustained interest, enabling them to make genuine and original contributions to discourse. While predicting whether a user is an expert on a given topic imposes strong constraints on who is a true positive, proficiency modeling implies a graded scoring, relaxing these constraints. Put another way, many active social media users can be assumed to possess, or eventually acquire, some level of proficiency on topics relevant to their community. We tackle proficiency modeling in an unsupervised manner by utilizing user embeddings to model engagement with a given topic, as indicated by a user's preference for authoring related content. We investigate five alternative approaches to model proficiency, ranging from basic ones to an advanced, tailored user modeling approach, applied within two real-world benchmarks for evaluation.

show abstract

Exploring Self-Identified Counseling Expertise in Online Support Forums

Cited by 11 publications

References 37 publications

Mitigating Toxic Degeneration with Empathetic Data: Exploring the Relationship Between Toxicity and Empathy

Mitigating Toxic Degeneration with Empathetic Data: Exploring the Relationship Between Toxicity and Empathy

A Shoulder to Cry on: Towards A Motivational Virtual Assistant for Assuaging Mental Agony

Modeling Proficiency with Implicit User Representations

Contact Info

Product

Resources

About