عربي English Türkçe اردو فارسى
Logo
Latest News
Varieties

New Study Evaluates AI Recognition of Lexical Borrowing

This Tuesday, Dec. 9, 2014 photo shows the word "culture" in the Merriam-Webster's Collegiate Dictionary, in New York. Merriam-Webster has named "culture" its 2014 word of the year. (AP Photo/Richard Drew)
This Tuesday, Dec. 9, 2014 photo shows the word "culture" in the Merriam-Webster's Collegiate Dictionary, in New York. Merriam-Webster has named "culture" its 2014 word of the year. (AP Photo/Richard Drew)
TT
TT

New Study Evaluates AI Recognition of Lexical Borrowing

This Tuesday, Dec. 9, 2014 photo shows the word "culture" in the Merriam-Webster's Collegiate Dictionary, in New York. Merriam-Webster has named "culture" its 2014 word of the year. (AP Photo/Richard Drew)
This Tuesday, Dec. 9, 2014 photo shows the word "culture" in the Merriam-Webster's Collegiate Dictionary, in New York. Merriam-Webster has named "culture" its 2014 word of the year. (AP Photo/Richard Drew)

Researchers from the Pontificia Universidad Católica del Perú (PUCP) and the Max Planck Institute for the Science of Human History have investigated the ability of machine learning algorithms to identify lexical borrowings, according to the German News Agency.

Lexical borrowing, or the direct transfer of words from one language to another, helps researchers trace the evolution of modern languages and indicate cultural contact between distinct linguistic groups. However, researchers often face challenges in this field because the tracing process requires the comparison of multiple languages.

"The automated detection of lexical borrowings is still one of the most difficult tasks we face in computational historical linguistics," the Phys.org website quoted lead author Johann-Mattis as saying.

In the current study, researchers trained language models that mimic the way in which linguists identify borrowings using acoustics to detect the words pronounced in the same way in different languages. This similarity indicates that the studied term was actually transferred from a language to another during the different phases of language evolution.

The team said the models were applied to a modified version of the World Loanword Database, a catalog of borrowing information for a sample of 40 languages from different language families all over the world, in order to see how accurately these models can determine the words borrowed from other languages.

In many cases the results were unsatisfying, suggesting that loanword detection is too difficult for machine learning methods most commonly used.

"After these first experiments with monolingual lexical borrowings, we can proceed to stake out other aspects of the problem," says researcher John Miller of PUCP.

Other researchers including co-author Tiago Tresoldi believe that "our computer-assisted approach will shed a new light on the importance of computer-assisted methods for language comparison and historical linguistics."

Most Viewed

 
 
Varieties

In Beirut, a Photographer's Frozen Moments Slow Down Time and Allow the Contemplation of Destruction

A bomb dropped from an Israeli jet hits a building in Ghobeiri, Beirut, Lebanon, Tuesday, October 22, 2024. (AP Photo/ Bilal Hussein)
A bomb dropped from an Israeli jet hits a building in Ghobeiri, Beirut, Lebanon, Tuesday, October 22, 2024. (AP Photo/ Bilal Hussein)
TT
TT

In Beirut, a Photographer's Frozen Moments Slow Down Time and Allow the Contemplation of Destruction

A bomb dropped from an Israeli jet hits a building in Ghobeiri, Beirut, Lebanon, Tuesday, October 22, 2024. (AP Photo/ Bilal Hussein)
A bomb dropped from an Israeli jet hits a building in Ghobeiri, Beirut, Lebanon, Tuesday, October 22, 2024. (AP Photo/ Bilal Hussein)

We watch video after video, consuming the world on our handheld devices in bites of two minutes, one minute, 30 seconds, 15. We turn to moving pictures — “film” — because it comes the closest to approximating the world that we see and experience. This is, after all, 2024, and video in our pocket — ours, others', everyone's — has become our birthright.
But sometimes — even in this era of live video always rolling, always recording, always capturing — sometimes the frozen moment can enter the eye like nothing else. And in the process, it can tell a larger story that echoes long after the moment was captured. That's what happened this past week in Beirut, through the camera lens of Associated Press photographer Bilal Hussein and the photographs he captured.
When Hussein set up his camera outside an evacuated Beirut apartment building Tuesday after Israel announced it would be targeted as part of military operations against Hezbollah, he had one goal in mind — only one. "All I thought of," he says, “was photographing the missile while it was coming down.”
He found a safe spot. He ensured a good angle. He wasn't stressed, he said; like many photographers who work in such environments, he had been in situations like this one before. He was ready.
When the attack came — a bomb, not a missile in the end — Hussein swung into action. And, unsurprisingly for a professional who has been doing this work for two decades, he did exactly what he set out to do.
Time slowed down
The sequence of images he made bursts with the explosive energy of its subject matter.
In one frame, the bomb hangs there, a weird and obtrusive interloper in the scene. It is not yet noticed by anyone around it, ready to bring its destruction to a building that, in moments, will no longer exist. The building's balconies, a split-second from nonexistence, are devoid of people as the bomb finds its mark.
These are the kind of moments that video, rolling at the speed of life or even in slow motion, cannot capture in the same way. A photo holds us in the scene, stops time, invites a viewer to take the most chaotic of events and break it down, looking around and noticing things in a strangely silent way that actual life could not.
In another frame, one that happened micro moments after the first, the building is in the process of exploding. Let's repeat that for effect, since even as recently as a couple generations ago photographs like this were rare: in the process of exploding.
Pieces of building are shooting out in all directions, in high velocity — in real life. But in the image they are frozen, outward bound, hanging in space awaiting the next seconds of their dissolution — just like the bomb that displaced them was doing milliseconds before. And in that, a contemplation of the destruction — and the people it was visited upon — becomes possible.
Tech gives us new prisms to see the world
The technology to grab so many images in the course of little more than one second — and do it in such clarity and high resolution — is barely a generation old.
So to see these “stills,” as journalists call them, come together to paint a picture of an event is a combination of artistry, intrepidity and technology — an exercise in freezing time, and in giving people the opportunity to contemplate for minutes, even hours, what took place in mere seconds. This holds true for positive things that the camera captures — and for visitations of violence like this one as well.
Photography is random access. We, the viewers of it, choose how to see it, process it, digest it. We go backward and forward in time, at will. We control the pace and the speed at which dizzying images hurtle at us. And in that process, something unusual for this era emerges: a bit of time to think.
That, among many other things, is the enduring power of the still image in a moving-picture world — and the power of what Bilal Hussein captured on that clear, sunny day in Beirut.