MA2---Towards-real-time-support-for-a-suicide-prevention-hotline-operators-

Challenging times such as the current COVID-19 pandemic can have a major impact on mental health1. One of the possible consequences of this downward trend in mental health is an increase in the prevalence of both mental illnesses such as depression and suicidal ideation. In the Netherlands, 113 suicide prevention runs the national suicide prevention helpline. Help seekers can reach out to the helpline by calling in, but they can also start to chat with one of the 113 volunteers. In the help-line at 113, the volunteers apply motivational interviewing2 during their conversations with the help seekers. At this time, 113 uses previous conversations as an educative tool to improve their expertise in the application of motivational interviewing. However, with this approach many chat sessions are never processed. To address this gap, lexical analysis of all chat sessions could be used analysis as an instrument to detect the cognitive and lexical markers of suicidal ideation and intent, raising the prospect of improving the effectiveness of suicide hotlines with interventions targeted at cognitive distortions associated with suicidal ideation. Examples of such analysis include sentiment analysis, e.g. ANEW3 or LIWC4, the annotation of cognitive distortions5 (as defined in cognitive behavioural therapy6), and the annotation of the effects of motivational interviewing2. The goal of this project is to track emotional and therapeutic/psychological features in the chat sessions that are recorded at the 113 helpline and use this information to provide the operators at the helpline with real-time feedback. With this feedback, you will help the operators assist the help seekers in their efforts to help each individual that reaches out to the suicide prevention hotline.