Scientists develop algorithm that predicts date of death: I'll find out my death date, but let's chat a little first
Scientists have developed an algorithm that predicts how and how long a person will live based on their life story. The model has a 78 percent accuracy rate and was trained using data from over 6 million people in Denmark. It is believed that the model could be used to identify factors that could lead to premature death and to reduce risk factors for a longer life.
12punto
Scientists have announced that they have developed an algorithm capable of predicting a person's date of death and how much money they will earn using a system they call 'life2vec'. This model, which works like chatbots, claims to have a 78 percent accuracy rate.
Developed by Danish and American scientists, life2vec fed machine learning algorithms using a vast data pool. This algorithm included information such as income, occupation, place of residence, disabilities, and pregnancy history of over 6 million people.
The model allows users to ask questions similar to chatbots, such as the probability of premature death for a specific person or income projections over a lifetime. The researchers stated that the model is not open to the general public or companies and that the personal information used is kept confidential to protect privacy laws.
It was emphasized that even if the model were to be made public, Danish privacy laws could restrict the use of life2vec to make decisions about individuals.