Scientists from Denmark claim to have developed a method for predicting a person's death using artificial intelligence. Their neural network Life2vec demonstrates an accuracy of nearly 80 percent in determining life expectancy. The results of the study were published in the journal Nature Computational Science.
“When creating the neural network, a person's life was analyzed as a language and text. By using knowledge of the life experiences of others and identifying patterns, it is possible to predict the future with high accuracy. This artificial intelligence, by carefully studying the details of people's lives, attempts to determine how much time they have left. At the same time, this application demonstrates how far we have advanced in technology and its impact on our perception of life and health,” says the study's author, Sune Lehmann from the Technical University of Denmark.
Data from various archives covering nearly 6 million residents of Denmark were used to train the neural network. According to the researchers, the algorithm achieved a death prediction accuracy of 79 percent.
During the research, scientists found that the life expectancy of wealthy individuals and those in managerial positions is higher than that of low-skilled workers. It was also established that people with mental disorders die earlier.
“These data are unlikely to be universal for residents of all countries. Since Life2vec was trained on Danish data, it will not be able to accurately predict life spans for populations of other states,” the developers warned.
Leave a comment