Using AI to Detect Family Violence Risk
Author Information
Author(s): Mavragani Amaryllis, Chrimes Dillon, Abd El-Hafeez Tarek, de Boer Kathleen, Mackelprang Jessica L, Nedeljkovic Maja, Meyer Denny, Iyer Ravi
Primary Institution: Swinburne University of Technology
Hypothesis
Can AI models accurately identify individuals at risk of family violence using textual or voice data?
Conclusion
This study aims to clarify how accurately AI models can identify individuals at high risk of perpetrating family violence.
Supporting Evidence
- Family violence is a major public health issue.
- AI has potential in detecting risks of family violence through voice and text analysis.
- Current prevention strategies for family violence have shown mixed success.
Takeaway
The study is looking at how well computers can tell if someone might hurt their family by listening to their voice or reading their words.
Methodology
The study will review existing research on AI's accuracy in detecting family violence risk using voice and text data.
Potential Biases
Potential biases in included studies may affect the overall findings.
Limitations
The search for studies has not yet begun, and results are not available.
Participant Demographics
The review will focus on adult human participants.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website