What is this algorithmic “death calculator” developed by Danish researchers that predicts life stages?

“It’s a very general framework for making predictions about human life. It can predict any training data.”, explains AFP Sune Lehmann, a professor at the Technical University of Denmark (DTU) and one of the authors of the study published in the journal Nature Computational Science. According to him, the possibilities are endless.

“It can predict health outcomes. So it can predict fertility or obesity or maybe who will get cancer or not. But it can also predict whether or not you’re going to make a lot of money.”He adds.

Specifically, life2vec uses the same operating model as ChatGPT but instead of processing textual data, it analyzes life stages such as birth, education, social benefits or even work hours.

“From a certain point of view, life is just a series of events: people are born, go to the pediatrician, go to school, move, get married, etc.

“, according to the study.

“Here we use this analogy to adapt an innovation in natural language processing to investigate the evolution and predictability of human life based on a detailed sequence of events.”She clarifies.

Six million pieces of data

It is based on anonymized data from nearly six million Danes collected by the National Statistics Institute. Sequence analysis makes it possible to predict the rest until the end. On death, the algorithm is correct in 78% of cases, on migration, in 73%.

“With a very small group of people between the ages of 35 and 65, we try to predict, based on the eight-year period from 2008 to 2016, whether a person will die in the next four years, up to 2020. The model pretty much does this. Well, Better than any other algorithm”explains Mr. Lehman, who is careful not to apply his formula to individual cases.

This age group, where deaths are generally low, allows the program’s reliability to be tested, according to the researchers.

But the tool is not ready to be used by common people as it still has biases. “For the moment, it’s a research project that explores the realm of possibilities (…), we don’t know if it treats everyone equally”.

It also remains to explore their long-term role, social connections, and impact on life expectancy.

counterweight

For academics, the project presents a scientific counterweight to the algorithms developed by Gafam (Google, Apple, Facebook, Amazon, Microsoft). “They may even build models like this, but they don’t publicize them, they don’t talk about them,” he says. “We can only hope that they develop us to buy more products”adds the researcher.

For him, it is “It’s important to have a public and open counterweight to understanding what we can do with this kind of data.”.

Especially since these types of algorithms are certainly already used in the insurance sector, says data ethics expert Pernil Tranberg.

“We are definitely put into groups (…) and this can be used against us to the extent that it can force us to pay higher insurance policies, restrict us from bank loans or access to public care because we are will die”She gives the list.

Bias is absent from the research project, which is not intended for personal use, thanks to the anonymization of its sources.

“No instances of data leaks” Personal with the National Statistics Institute, and “Data is not personal”, she emphasizes. However, with the development of artificial intelligence, “Everything’s Going Fast”.

project “Just shows that we have a lot of data in Denmark and can use it because we are all going in the same direction”, Ms. adds Tranberg. And some developers have decided to use this idea for commercial purposes. “On the web, we already see prediction clocks, which show how old we will be and some are not at all reliable”

she warns.