Another AI capability has been discovered by researchers, and while it’s interesting, it might not be as funny as other tech-related findings.
According to a recent study published in Nature Computational Science, among other life details, a novel artificial intelligence system that approaches human lives like language might be able to forecast death within a specific time frame.
A machine-learning model known as life2vec was created by Danish researchers as part of the study. It has the ability to forecast a person’s life facts, such as death, foreign relocation, and personality attributes.
Millions of citizens’ birth dates, sexes, jobs, locations, and usage of the nation’s universal healthcare program are among the data points incorporated into the model.
The approach outperformed other predictive models in forecasting death over a four-year period, with over 78% accuracy.
Life2vec shown encouraging initial results in relating personality qualities to life events, as seen by its 73% accuracy rate in predicting individuals’ departure from Denmark and their self-reported answers to a personality questionnaire in a different test.
According to computational social science researcher Matthew Salganik, a professor of sociology at Princeton University, the work shows an innovative new method for forecasting and analyzing people’s life trajectories.
As far as he knows, no one has ever used the “very different style” that the life2vec developers employ.
Lehmann and his colleagues created life2vec, a language processing tool that creates personalized timelines of events such as hospital admissions and wage fluctuations based on each person’s data.
Life2vec is a potential tool for future prediction because of its flexible model architecture, which makes it easy to adjust and fine-tune and offers predictions about many undiscovered elements of human existence.
Medical experts have already gotten in touch with Lehmann to request his assistance in creating health-related versions of life2vec, such as one that might help identify population-level risk factors for uncommon diseases.
He intends to use a tool to examine the impact of relationships on quality of life and compensation, as well as to reveal hidden societal prejudices. Examples of these biases include surprising correlations between age or place of birth and professional progress.
