A Computer Just Read 25 Million Articles From 150 Years Of News, And Taught Us This

The results offer a more accurate insight into the past

The Evening Standard, 1866. Image Evening Standard The British Newspaper Archive Blog

“Study the past if you would define the future,” said Confucius. 

Never have truer words been spoken – it goes without saying that the study of our past is something we as a human race obsess over.

Researchers at the University of Bristol have taken it a step further, entering 25 million news articles British publications from the last 150 years into a computer to investigate our social habits over time. The study was published in Proceedings of the National Academy of Sciences of the United States of America.

The primary focus of the experiment is ‘to extract macroscopic trends in history and culture, including gender bias, geographical focus, technology, and politics, along with accurate dates for specific events’, write the team of authors in the study titled Content Analysis of 150 years of British Periodicals.


A report from 3rd September 1939. Nottingham Evening Post Image Nottingham Evening Post


Extract they did, finding a host of patterns in the broad landscape of reports. One such trend was the increase of women in the news over the 20th century. Before this era, there was a significant ‘underrepresentation’ of female-driven articles.

Another, highlighted by New Atlas, was the discovery of just who attracts the most fame over time. It seems that writers and politicians are more likely to be remembered by history, while scientists and mathematicians fall short. Sorry nerd friends.

Even with such results, the team concluded that nothing beats the human thought process, saying:

“What cannot be automated is the understanding of the implications of these findings for people, which will always be the realm of the humanities and social sciences and will never be that of machines.”

Previous Post
Next Post