首页|Research Department Reports Findings in Machine Learning (The Most Effective Int erventions for Classification Model Development to Predict Chat Outcomes Based o n the Conversation Content in Online Suicide Prevention Chats: Machine Learning ...)
Research Department Reports Findings in Machine Learning (The Most Effective Int erventions for Classification Model Development to Predict Chat Outcomes Based o n the Conversation Content in Online Suicide Prevention Chats: Machine Learning ...)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Machine Learning is th e subject of a report. According to news reporting originating from Amsterdam, N etherlands, by NewsRx correspondents, research stated, "For the provision of opt imal care in a suicide prevention helpline, it is important to know what contrib utes to positive or negative effects on help seekers. Helplines can often be con tacted through text-based chat services, which produce large amounts of text dat a for use in large-scale analysis." Our news editors obtained a quote from the research from Research Department, "W e trained a machine learning classification model to predict chat outcomes based on the content of the chat conversations in suicide helplines and identified th e counsellor utterances that had the most impact on its outputs. From August 202 1 until January 2023, help seekers (N=6903) scored themselves on factors known t o be associated with suicidality (eg, hopelessness, feeling entrapped, will to l ive) before and after a chat conversation with the suicide prevention helpline i n the Netherlands (113 Suicide Prevention). Machine learning text analysis was u sed to predict help seeker scores on these factors. Using 2 approaches for inter preting machine learning models, we identified text messages from helpers in a c hat that contributed the most to the prediction of the model. According to the m achine learning model, helpers' positive affirmations and expressing involvement contributed to improved scores of the help seekers. Use of macros and ending th e chat prematurely due to the help seeker being in an unsafe situation had negat ive effects on help seekers. This study reveals insights for improving helpline chats, emphasizing the value of an evocative style with questions, positive affi rmations, and practical advice."
AmsterdamNetherlandsEuropeCyborgsEmerging TechnologiesHealth and MedicineMachine LearningMental HealthRi sk and PreventionSuicide
NETL
NSTL
