Developing conversational Virtual Humans for social emotion elicitation based on large language models

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 246, P. 123261 - 123261

Published: Jan. 18, 2024

Emotions play a critical role in numerous processes, including, but not limited to, social interactions. Consequently, the ability to evoke and recognize emotions is challenging task with widespread implications, notably field of mental health assessment systems. However, up until now, emotional elicitation methods have utilized simulated open conversations. Our study introduces comprehensive Virtual Human (VH), equipped realistic avatar conversational abilities based on Large Language Model. This architecture integrates psychological constructs—such as personality, mood, attitudes—with facial expressions, lip synchronization, voice synthesis. All these features are embedded into modular, cognitively-inspired framework, specifically designed for voice-based semi-guided conversations real time. The validation process involved an experiment 64 participants interacting six distinct VHs, each provoke different basic emotion. system took average 4.44 s generate VH's response. Participants assessed naturalness realism conversation, scoring averages 4.61 out 7, respectively. VHs successfully generated intended valence users, while arousal was evoked, though it could be recognized VHs. findings underscore feasibility employing within affective computing elicit socially ecologically valid contexts. development holds significant potential application sectors such health, education, marketing, among others.

Language: Английский

---

Achieving health equity through conversational AI: A roadmap for design and implementation of inclusive chatbots in healthcare

PLOS Digital Health, Journal Year: 2024, Volume and Issue: 3(5), P. e0000492 - e0000492

Published: May 2, 2024

Background The rapid evolution of conversational and generative artificial intelligence (AI) has led to the increased deployment AI tools in healthcare settings. While these promise efficiency expanded access services, there are growing concerns ethically, practically terms inclusivity. This study aimed identify activities which reduce bias make their designs implementation more equitable. Methods A qualitative research approach was employed develop an analytical framework based on content analysis 17 guidelines about use clinical stakeholder consultation subsequently conducted with a total 33 ethnically diverse community members, designers, industry experts relevant health professionals further roadmap for equitable design healthcare. Framework interview data. Results 10-stage developed outline phases: 1) Conception planning, 2) Diversity collaboration, 3) Preliminary research, 4) Co-production, 5) Safety measures, 6) testing, 7) Healthcare integration, 8) Service evaluation auditing, 9) Maintenance, 10) Termination. Discussion We have made specific recommendations increase AI’s equity as part services. These emphasise importance collaborative involvement patient groups navigating technologies. Further must assess impact recommended chatbots’ fairness ability inequalities.

Language: Английский

---

Health Chatbots for Fighting COVID-19: a Scoping Review

Acta Informatica Medica, Journal Year: 2020, Volume and Issue: 28(4), P. 241 - 241

Published: Jan. 1, 2020

Background: Health chatbots are rising in popularity and capability for fighting the novel SARS-CoV-2 coronavirus (COVID-19). Objectives: This study aims to review current literature on COVID-19 related healthcare, identify characterize these emerging technologies their applications combating COVID-19, describe challenges. Methods: The authors conducted a scoping of peer-reviewed guided by Arksey O'Malley framework. PubMed/MEDLINE Google Scholar were searched over period between January September 2020 using keywords "COVID* chatbot", "virtual assistant", "AI enabled platform COVID" associated synonyms. Relevant studies' references checked further articles. content studies was screened thematically analyzed two authors. Results: Out 543 articles initially identified, 9 eligible inclusion. Studies describing chatbots' development architecture (n=6) most common, only 3 empirical user experience identified. Our identified five key health chatbots, which were: disseminating information knowledge; self-triage personal risk assessment; monitoring exposure notifications; tracking symptoms aspects; misinformation fake news. Furthermore, can accomplish following tasks: ask answer questions; create records history use; complete forms generate reports; take simple actions. Nonetheless, use poses many challenges both at level social system (i.e., consumers' acceptability) as well technical design usability). Conclusion: Using combat is practice still its infancy. We believe that our work will help researchers this domain gain better understanding technology's applications, needed continuous improvement functionalities usefulness fight COVID-19.

Language: Английский

---

Conversational Agents: Goals, Technologies, Vision and Challenges

Sensors, Journal Year: 2021, Volume and Issue: 21(24), P. 8448 - 8448

Published: Dec. 17, 2021

In recent years, conversational agents (CAs) have become ubiquitous and are a presence in our daily routines. It seems that the technology has finally ripened to advance use of CAs various domains, including commercial, healthcare, educational, political, industrial, personal domains. this study, main areas which successful described along with technologies enable creation CAs. Capable conducting ongoing communication humans, encountered natural-language processing, deep learning, integrate emotional aspects. The used for evaluation publicly available datasets outlined. addition, several future research identified address moral security issues, given current state CA-related technological developments. uniqueness review is an overview concepts building blocks provided, categorized according their abilities application primary tools may be useful development different categories described. Finally, some thoughts directions domains benefit from introduced.

Language: Английский

---

Voice-Based Conversational Agents for the Prevention and Management of Chronic and Mental Health Conditions: Systematic Literature Review

Journal of Medical Internet Research, Journal Year: 2021, Volume and Issue: 23(3), P. e25933 - e25933

Published: March 3, 2021

Background Chronic and mental health conditions are increasingly prevalent worldwide. As devices in our everyday lives offer more voice-based self-service, conversational agents (VCAs) have the potential to support prevention management of these a scalable manner. However, evidence on VCAs dedicated chronic is unclear. Objective This study provides better understanding current methods used evaluation interventions for delivered through VCAs. Methods We conducted systematic literature review using PubMed MEDLINE, Embase, PsycINFO, Scopus, Web Science databases. included primary research involving or VCA reporting an empirical system either terms accuracy, technology acceptance, both. A total 2 independent reviewers screening data extraction, agreement between them was measured Cohen kappa. narrative approach synthesize selected records. Results Of 7170 prescreened papers, 12 met inclusion criteria. All studies were nonexperimental. The provided behavioral (n=5), monitoring services (n=3), both (n=4). via smartphones tablets (n=2), smart speakers (n=3). In cases, no device specified. 3 targeted cancer, whereas diabetes heart failure. other hearing impairment, asthma, Parkinson disease, dementia, autism, intellectual disability, depression. majority (n=7) assessed but only few (n=3) validated instruments. Half (n=6) reported performance measures speech recognition ability respond health-related queries. Only minority (n=2) measure attitudes toward intervention-targeted behavior. Moreover, (n=4) controlling participants’ previous experience with technology. Finally, risk bias varied markedly. Conclusions heterogeneity methods, limited number identified, high show that still its infancy. Although results accuracy acceptance encouraging, there need establish conclusive efficacy conditions, absolute comparison standard care.

Language: Английский

---

Conversational Agents as Mediating Social Actors in Chronic Disease Management Involving Health Care Professionals, Patients, and Family Members: Multisite Single-Arm Feasibility Study

Journal of Medical Internet Research, Journal Year: 2021, Volume and Issue: 23(2), P. e25060 - e25060

Published: Jan. 23, 2021

Successful management of chronic diseases requires a trustful collaboration between health care professionals, patients, and family members. Scalable conversational agents, designed to assist may play significant role in supporting this scalable way by reaching out the everyday lives patients their However, date, it remains unclear whether such role, would be accepted they can support multistakeholder collaboration.

Language: Английский

---

Health Apps for Combating COVID-19: Descriptive Review and Taxonomy

JMIR mhealth and uhealth, Journal Year: 2021, Volume and Issue: 9(3), P. e24322 - e24322

Published: Feb. 22, 2021

Mobile phone apps have been leveraged to combat the spread of COVID-19. However, little is known about these technologies' characteristics, technical features, and various applications in health care when responding this public crisis. The lack understanding has led developers governments make poor choices apps' designs, which resulted creating less useful that are overall appealing consumers due their flaws.This review aims identify, analyze, categorize related COVID-19 currently available for app stores; particular, it focuses on exploring key features classifying purposes were designed serve.A was conducted using PRISMA-ScR (Preferred Reporting Items Systematic Reviews Meta-Analyses Extension Scoping Reviews) guidelines. Apple Store Google Play searched between April 20 September 11, 2020. An included if dedicated disease listed under medical categories stores. descriptions extracted from web pages thematically analyzed via open coding identify both purpose. characteristics summarized presented with descriptive statistics.Of 298 initially retrieved, 115 met inclusion criteria. A total 29 found our sample apps, then categorized into five 77 (67%) developed by or national authorities purpose promoting users track personal (9/29, 31%). Other raising awareness how (8/29, 27%), managing exposure (6/29, 20%), monitoring professionals (5/29, 17%), conducting research studies (1/29, 3.5%).This study provides an overview taxonomy market based differences basic As most provided authorities, indicates essential role as tools crisis management. By involving population self-tracking providing them technology self-assess, deemed be a driver participatory approach curtail Further effort required researchers evaluate effectiveness governmental organizations increase digital solutions.

Language: Английский

---

See you soon again, chatbot? A design taxonomy to characterize user-chatbot relationships with different time horizons

Computers in Human Behavior, Journal Year: 2021, Volume and Issue: 127, P. 107043 - 107043

Published: Oct. 5, 2021

Users interact with chatbots for various purposes and motivations – different periods of time. However, since are considered social actors given that time is an essential component interactions, the question arises as to how need be designed depending on whether they aim help individuals achieve short-, medium- or long-term goals. Following a taxonomy development approach, we compile 22 empirically conceptually grounded design dimensions contingent chatbots' temporal profiles. Based upon classification analysis 120 therein, abstract three time-dependent chatbot archetypes: Ad-hoc Supporters, Temporary Assistants, Persistent Companions. While serves blueprint researchers designers developing evaluating in general, our archetypes also offer practitioners academics alike shared understanding naming convention study

Language: Английский

---

An Artificial Intelligence Chatbot for Young People’s Sexual and Reproductive Health in India (SnehAI): Instrumental Case Study

Journal of Medical Internet Research, Journal Year: 2022, Volume and Issue: 24(1), P. e29969 - e29969

Published: Jan. 3, 2022

Leveraging artificial intelligence (AI)-driven apps for health education and promotion can help in the accomplishment of several United Nations sustainable development goals. SnehAI, developed by Population Foundation India, is first Hinglish (Hindi + English) AI chatbot, deliberately designed social behavioral changes India. It provides a private, nonjudgmental, safe space to spur conversations about taboo topics (such as sex family planning) offers accurate, relatable, trustworthy information resources.This study aims use Gibson theory affordances examine SnehAI offer scholarly guidance on how chatbots be used educate adolescents young adults, promote sexual reproductive health, advocate entitlements women girls India.We adopted an instrumental case approach that allowed us explore from perspectives technology design, program implementation, user engagement. We also mix qualitative insights quantitative analytics data triangulate our findings.SnehAI demonstrated strong evidence across fifteen functional affordances: accessibility, multimodality, nonlinearity, compellability, queriosity, editability, visibility, interactivity, customizability, trackability, scalability, glocalizability, inclusivity, connectivity, actionability. effectively engaged its users, especially men, with 8.2 million messages exchanged 5-month period. Almost half incoming were texts deeply personal questions concerns well allied topics. Overall, successfully presented itself trusted friend mentor; curated content was both entertaining educational, natural language processing system worked personalize chatbot response optimize experience.SnehAI represents innovative, engaging, educational intervention enables vulnerable hard-to-reach population groups talk learn sensitive important issues. powerful testimonial vital potential lies technologies good.

Language: Английский

---

Ask Rosa – The making of a digital genetic conversation tool, a chatbot, about hereditary breast and ovarian cancer

Patient Education and Counseling, Journal Year: 2021, Volume and Issue: 105(6), P. 1488 - 1494

Published: Oct. 6, 2021

We aimed at developing a pilot version of an app (Rosa) that can perform digital conversations with breast or ovarian cancer patients about genetic BRCA testing, using chatbot technology, to identify best practices for future patient-focused chatbots.We chose commercial platform and participatory methodology team patient representatives, IT engineers, counselors clinical geneticists, within nationwide collaboration. An iterative approach ensured extensive user formal usability testing during the development process.The phase lasted two years until was completed in December 2019. The iteration steps disclosed major challenges artificial intelligence (AI)-based matching provided questions predefined information database, leading initially high level fallback answers. therefore developed strategies reduce potential language ambiguities (e.g. BRCA1 vs BRCA2) overcome dialogue confusion. first prototype contained database 500 67 corresponding answers, while final included 2257 144 Despite limited AI functionality chatbot, revealed users liked layout found trustworthy reader friendly.Building health is challenging, expensive time consuming today's technology. had positive attitude would use it real life setting, if given them by care personnel.We here present framework initiatives. combination perspective incorporated every process. strongly recommend this patient-centered innovations.

Language: Английский

---