Doom Calculator 'Life2vec' Algorithm Can Predict Person's Earnings, Time of Death With 78% Accuracy

A novel algorithm developed by scientists makes use of a person's life story to predict earnings and the time of death.

Life2Vec: Doom Calculator Algorithm

Based on the new study entitled "Using sequences of life-events to predict human lives" on the doom calculator, the algorithm called Life2vec, boasts 78% accuracy. This makes it on par with different algorithms that were developed to predict life outcomes that are similar.

However, Life2vec functions like a chatbot, similar to ChatGPT, which is unlike how other models work. It maximizes existing details to predict what could happen in such aspects of the future.

The development of the bot is attributed to scientists from the US and Denmark. These specialists trained the algorithm, which works on machine learning, with a huge Danish data pool. The algorithm was fed with different data from more than six million real individuals. The data, which was from 2008 to 2016, included residence area, profession, income, injuries, and even pregnancy history.

As a result, the model was capable of processing plain language and also making predictions of how likely a person is to die early and how much the person could earn throughout the lifespan. It makes predictions of where one's story may go based on the written information provided.

Some factors linked to earlier death were having a mental health diagnosis, being male, or having a skilled profession. Things that were associated with a longer lifespan were having a leadership role and a higher income, among others.

Based on the data of the population, the algorithm was able to correctly predict who would pass away by 2020 with more than three-fourths accuracy.

Sune Lehmann, the lead researcher, explains that similar technologies for life event and human behavior predictions have already been incorporated today with tech firms that can perform different functions, such as tracking social network behavior, conducting extremely accurate profiling, and using such profiles for predicting behavior and influencing it.

With this, it is crucial for the discussion to be included in democratic conversations in order to see where the trajectory of technology leads to and whether this development is something that humanity wants to embrace.

Not Publicly Available

To protect the information of the people behind the system training, the algorithm is not available to the public. Lehmann explains that they are working to find ways to share results more openly. However, further research is necessary for this to be done in a manner that can guarantee that the people involved in the study will have protected privacy.

Should the model be made available to the general public, privacy laws in Denmark will restrict the use of life2vec for making decisions about people, such as hiring decisions or making insurance policies.

Check out more news and information on Tech & Innovation in Science Times.