This is something that researchers at the University of Southern California are looking into, and they are using none other than wearable gadgets to help them make their predictions. How this works is that researchers gave couples a wearable that can track biodata like body temperature, heart activity, and sweat. Participants in the research were also required to log their feelings towards their partner every hour, which also measured things like content and intensity of speech.
After that through the use of machine learning, researchers were able to capture moments of conflicts with an accuracy rating as high as 86%. According to the researchers, the next step in the process would be to create a model that could predict conflicts up to five minutes before they occur, which in turn could allow device makers to focus not just on physical health, but emotional health as well.
Speaking to TechCrunch, the study’s co-author Theodora Chaspari said, “The fact that we are capturing bio-signals from wearables – this is a source of information we can get from people that we cannot see with the naked eye. It was a pretty useful source of information.”