Science & Technology

A new way to predict the future with machines that read the present

I usually want to talk about reading history and predicting the future.

But now that I’ve found two recent articles, I had to think about the latest publications that would help set the course for future science. Think of Google, but it’s not that evil as it has nothing to do with advertising.

Initially The story of the “giant” It has published an index of 107 million papers in a clever way to navigate current copyright law.

Some researchers who have early access to the index say it is a major step forward in helping them search the literature using software. This is a procedure called text mining. Gitanjali Yadav, a computational biologist at the University of Cambridge, UK, who studies volatile organic compounds released from plants, examines Malamud’s index to analyze phytochemicals found in research papers around the world. He states that he is aiming for that. “There is no way for me, or anyone else, to experimentally analyze or measure the chemical fingerprints of all plant species on Earth. Much of the information we seek is already present in the published literature. “I will,” she says. However, researchers are limited by lack of access to many papers, Yadav adds. Malamud’s “Comprehensive Index”, as he calls it, aims to address the problems faced by researchers like Yadav.

The second is a paper on predicting research trends using. Computational analysis of available papers..

Here we show the development of a semantic network of quantum physics called SEMNET, using 750,000 scientific papers and knowledge from books and Wikipedia. Used in combination with artificial neural networks to predict future research trends. Individual scientists can use SEMNET to propose and inspire personalized, ready-to-use ideas. Computer-inspired scientific ideas play an important role in accelerating scientific progress. We hope that our work will contribute directly to that important goal.

Source: PNAS, January 28, 2020, vol. 117, no. 4. “When two concepts appear simultaneously in any title or abstract of any of the 750,000 papers, an edge is formed.”

Calling Douglas Adams is always fascinating The famous “42” story When reading these types of articles.

The method used looks more mathematical and rushes to conclusions compared to what an experienced historian might do to test trends and meanings.

A new way to predict the future with machines that read the present A new way to predict the future with machines that read the present

Back to top button