scholarly journals Conversational Agents for Recipe Recommendation

2020 ◽  
Author(s):  
Sabrina Barko-Sherif ◽  
David Elsweiler ◽  
Morgan Harvey
Keyword(s):  
Conversational Agents

A fully automated conversational agent with artificial intelligence capabilities (Youper) for treating depression and anxiety: A nonrandomized feasibility trial (Preprint)

2019 ◽  
Author(s):  
Jose Hamilton Vargas ◽  
Thiago Antonio Marafon ◽  
Diego Fernando Couto ◽  
Ricardo Giglio ◽  
Marvin Yan ◽  
...  
Keyword(s):  
Mental Health ◽  
Artificial Intelligence ◽  
Financial Incentives ◽  
Empirical Support ◽  
Digital Technologies ◽  
Feasibility Trial ◽  
Depression And Anxiety ◽  
Conversational Agents ◽  
Symptoms Of Depression ◽  
Viable Solution

BACKGROUND Mental health conditions, including depression and anxiety disorders, are significant global concerns. Many people with these conditions don't get the help they need because of the high costs of medical treatment and the stigma attached to seeking help. Digital technologies represent a viable solution to these challenges. However, these technologies are often characterized by relatively low adherence and their effectiveness largely remains empirical unverified. While digital technologies may represent a viable solution for this persisting problem, they often lack empirical support for their effectiveness and are characterized by relatively low adherence. Conversational agents using artificial intelligence capabilities have the potential to offer a cost-effective, low-stigma and engaging way of getting mental health care. OBJECTIVE The objective of this study was to evaluate the feasibility, acceptability, and effectiveness of Youper, a mobile application that utilizes a conversational interface and artificial intelligence capabilities to deliver cognitive behavioral therapy-based interventions to reduce symptoms of depression and anxiety in adults. METHODS 1,012 adults with symptoms of depression and anxiety participated in a real-world setting study, entirely remotely, unguided and with no financial incentives, over an 8-week period. Participants completed digital versions of the 9-item Patient Health Questionnaire (PHQ-9) and the 7-item Generalized Anxiety Disorder scale (GAD-7) at baseline, 2, 4, and 8 weeks. RESULTS After the eight-week study period, depression (PHQ-9) scores of participants decreased by 48% while anxiety (GAD-7) scores decreased by 43%. The RCI was outside 2 standard deviations for 93.0% of the individuals in the PHQ-9 assessment and 90.7% in the GAD-7 assessment. Participants were on average 24.79 years old (SD 7.61) and 77% female. On average, participants interacted with Youper 0.9 (SD 1.56) times per week. CONCLUSIONS Results suggest that Youper is a feasible, acceptable, and effective intervention for adults with depression and anxiety. CLINICALTRIAL Since this study involved a nonclinical population, it wasn't registered in a public trials registry.

App review of anxiety and depression chatbots and their self-care features (Preprint)

2020 ◽  
Author(s):  
Arfan Ahmed ◽  
Nashva ALi ◽  
Sarah Aziz ◽  
Alaa A Abd-Alrazaq ◽  
Asmaa Hassan ◽  
...  
Keyword(s):  
Mental Health ◽  
Meta Analysis ◽  
Self Care ◽  
Anxiety And Depression ◽  
Conversational Agents ◽  
Eligibility Criteria ◽  
On Demand ◽  
User Ratings ◽  
Google Play

BACKGROUND Anxiety and depression rates are at an all-time high along with other mental health disorders. Smartphone-based mental health chatbots or conversational agents can aid psychiatrists and replace some of the costly human based interaction and represent a unique opportunity to expand the availability and quality of mental health services and treatment. Regular up-to-date reviews will allow medics and individuals to recommend or use anxiety and depression related smartphone based chatbots with greater confidence. OBJECTIVE Assess the quality and characteristics of chatbots for anxiety and depression available on Android and iOS systems. METHODS A search was performed in the App Store and Google Play Store following the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) protocol to identify existing chatbots for anxiety and depression. Eligibility of the chatbots was assessed by two individuals based on predefined eligibility criteria. Meta-data of the included chatbots and their characteristics were extracted from their description and upon installation by 2 reviewers. Finally, chatbots quality information was assessed by following the mHONcode principles. RESULTS Although around 1000 anxiety and depression related chatbots exist, only a few (n=11) contained actual chatbots that could provide the user a real substitute for a human-human based interaction, even with today's Artificial Intelligence advancements, only one of these chatbots had voice as an input/output modality. Of the selected apps that contained chatbots all were clearly built with a therapeutic human substitute goal in mind. The majority had high user ratings and downloads highlighting the popularity of such chatbots and their promising future within the realm of anxiety and depression. CONCLUSIONS Anxiety and depression chatbot apps have the potential to increase the capacity of mental health self-care providing much needed assistance to professionals. In the current covid-19 pandemic, chatbots can also serve as a conversational companion with the potential of combating loneliness, especially in lockdowns where there is a lack of social interaction. Due to the ubiquitous nature of chatbots users can access them on-demand at the touch of a screen on ones’ smartphone. Self-care interventions are known to be effective and exist in various forms and some can be made available as chatbot features, such as assessment, mood tracking, medicine tracking, or simply providing conversation in times of loneliness.

scholarly journals Comparison of disclosure/concealment of medical information given to conversational agents or to physicians

2021 ◽  
Vol 27 (1) ◽  
pp. 146045822199486
Author(s):  
Nicholas RJ Frick ◽  
Felix Brünker ◽  
Björn Ross ◽  
Stefan Stieglitz
Keyword(s):  
Natural Language ◽  
Medical Information ◽  
Frame Of Reference ◽  
Future Research ◽  
Negative Consequences ◽  
Conversational Agents ◽  
Online Questionnaire ◽  
Computer Based

Within the anamnesis, medical information is frequently withheld, incomplete, or incorrect, potentially causing negative consequences for the patient. The use of conversational agents (CAs), computer-based systems using natural language to interact with humans, may mitigate this problem. The present research examines whether CAs differ from physicians in their ability to elicit truthful disclosure and discourage concealment of medical information. We conducted an online questionnaire with German participants ( N = 148) to assess their willingness to reveal medical information. The results indicate that patients would rather disclose medical information to a physician than to a CA; there was no difference in the tendency to conceal information. This research offers a frame of reference for future research on applying CAs during the anamnesis to support physicians. From a practical view, physicians might gain better understanding of how the use of CAs can facilitate the anamnesis.

Closing the Loop in Computer Agent/Patient Communication

2020 ◽  
Vol 64 (1) ◽  
pp. 711-715
Keyword(s):  
Older Adults ◽  
Best Practice ◽  
Self Care ◽  
Effective Teacher ◽  
Care Providers ◽  
Conversational Agents ◽  
Patient Centered ◽  
Face To Face ◽  
Communication Techniques ◽  
Improve Instruction

While patient-centered communication supports patient self-care, providers rarely have enough time to consistently use patient- centered communication techniques. Technology has potential to support patient-centered communication, but frequently older adults prefer face-to-face communication with providers. Conversational agents (CAs) may support provider communication with older adults by emulating best practices from face-to-face communication. We investigated older adults’ response to a prototype CA communication system that served as a virtual provider and presented medication instructions using teachback, a recommended best practice that involves asking patients questions to ‘close the communication loop’. Older adults were told how to take medications by a CA who used (or did not use) teachback, and then were interviewed. Older adults were open to interacting with the CA and thought it would help support self-care. They thought the CA was a more effective teacher when using teachback and that this interactive strategy helped them remember the instructions. However, teachback did not improve instruction memory.

scholarly journals Attention-based hierarchical recurrent neural networks for MOOC forum posts analysis

2020 ◽  
Author(s):  
Nicola Capuano ◽  
Santi Caballé ◽  
Jordi Conesa ◽  
Antonio Greco
Keyword(s):  
Neural Networks ◽  
Recurrent Neural Networks ◽  
Text Categorization ◽  
Online Courses ◽  
Recurrent Network ◽  
Subject Area ◽  
Conversational Agents ◽  
Collaborative Tools ◽  
Experimental Findings ◽  
Productive Talk

AbstractMassive open online courses (MOOCs) allow students and instructors to discuss through messages posted on a forum. However, the instructors should limit their interaction to the most critical tasks during MOOC delivery so, teacher-led scaffolding activities, such as forum-based support, can be very limited, even impossible in such environments. In addition, students who try to clarify the concepts through such collaborative tools could not receive useful answers, and the lack of interactivity may cause a permanent abandonment of the course. The purpose of this paper is to report the experimental findings obtained evaluating the performance of a text categorization tool capable of detecting the intent, the subject area, the domain topics, the sentiment polarity, and the level of confusion and urgency of a forum post, so that the result may be exploited by instructors to carefully plan their interventions. The proposed approach is based on the application of attention-based hierarchical recurrent neural networks, in which both a recurrent network for word encoding and an attention mechanism for word aggregation at sentence and document levels are used before classification. The integration of the developed classifier inside an existing tool for conversational agents, based on the academically productive talk framework, is also presented as well as the accuracy of the proposed method in the classification of forum posts.

Sign in / Sign up

Export Citation Format

Share Document