Data Science

Jeannette Wing in a beige suit, in a white, well-lit room
The nomination will ensure equity, fairness, and accountability in algorithms consistent with the “data for good” mission of Columbia’s Data Science Institute.
Will Geary
As Will Geary leaves the Engineering School he’ll start executing his ideas at CitySwifter, an Ireland-based startup using deep learning to optimize city bus networks.
Jeannette Wing in a beige suit, in a white, well-lit room
From filter bubbles to fake news, data-driven algorithms have developed a reputation problem. Jeannette Wing, as the new director of Columbia’s Data Science Institute, has woven ethics and social impact into the Institute’s mission.
hydra image with neurons labeled with a green fluorescence indicator

Researchers show how an algorithm for filtering spam can learn to pick out, from hours of video footage, the full behavioral repertoire of tiny, pond-dwelling Hydra. In the above image, hydra's neurons are labeled with a green fluorescence indicator. (Yuste Lab, Columbia University)

Researchers show how an algorithm for filtering spam can learn to pick out, from hours of video footage, the full behavioral repertoire of tiny, pond-dwelling Hydra. By comparing Hydra’s behaviors to the firing of its neurons, the researchers hope to eventually understand how its nervous system, and that of more complex animals, works.
Instagram selfie of study coauthor Ana-Andreea Stoica

A network effect known as homophily may reduce women’s visibility on social media when recommendation algorithms are added, says a new study. Above, a selfie from study coauthor Ana-Andreea Stoica's Instagram account. (Courtesy of Stoica)

When recommendation algorithms are turned loose on a social network with homophily, women become less visible, says a new study by Columbia researchers.
Yaniv Erlich holds a framed picture of his father as a teenager

Columbia computer scientist Yaniv Erlich has assembled the world's largest family tree to date. He and his father (pictured as a teenager) are among the 86 million genealogy profiles Erlich and his colleagues drew from in their study. (MyHeritage)

From 86 million genealogy profiles, researchers have amassed the largest, scientifically-vetted family tree to date, which at 13 million people, is slightly bigger than a nation the size of Cuba or Belgium. Published in the journal Science, the new dataset offers fresh insights into the last 500 years of marriage and migration in Europe and North America, and the role of genes in longevity.

Technology is changing journalism in ways no one could have imagined when Columbia Journalism School opened in 1912, and the algorithms, bots and trolls that affect how we get news today are just some of the complicating factors in the new information ecosystem.

New technology adapted to cheap DNA sequencers can rapidly identify people and cells from their DNA.

New technology adapted to cheap DNA sequencers can rapidly identify people and cells from their DNA. Here, researcher Sophie Zaaijer demonstrates from a NYC rooftop how easy DNA-authentication can be.

Researchers have developed a method to quickly and accurately identify people and cell lines from their DNA. The technology, described in the latest issue of eLife, has a wide range of applications, but its most immediate use could be to flag mislabeled or contaminated cell lines in cancer experiments.
In a new essay in Nature, Columbia neuroscientist Rafael Yuste joins more than two dozen researchers in calling for ethical guidelines to cover the evolving use of computer hardware and software to enhance or restore human capabilities.
In a study analyzing the genomes of 210,000 people in the United States and Britain, Columbia researchers find that the genetic variants linked to Alzheimer’s disease and heavy smoking are less frequent in people with longer lifespans, suggesting that natural selection is weeding out these unfavorable variants in both populations.

Pages