A Columbia data scientist aims to develop tools to manage demand with flexible options for consumers, converting their flexibility into operational efficiency for companies. He plans to create mathematical frameworks to optimize pricing and scheduling, benefiting both consumers and firms.
A Columbia professor is using data science to design interventions and recommend policies that help the most vulnerable populations overcome inequalities in access and outcomes. His project, Nudging New York, aims to eliminate inequality and increase healthcare access in underserved communities.
A Columbia University research team has received a Facebook Probability and Programming research award to develop static analysis methods for probabilistic programming. The team aims to enhance the usability and accuracy of probabilistic programming by creating tools that automate model checking and compiler verification.
Researchers will create a risk dashboard to help ISOs make sound decisions quickly and allocate resources for reducing renewable energy risks, making the U.S. power grid more efficient and reliable.
The COVID-19 Trial Finder system generates user-specific questionnaires to filter out applicable trials, reducing a large set of potential trials to a handful within 5-6 survey questions.
A Columbia study published in Obstetrics & Gynecology found that patients using copper intrauterine devices (Cu IUDs) had a lower risk of high-grade cervical neoplasms compared to those using levonorgestrel-releasing intrauterine systems (LNG-IUS). The study analyzed data from over 10,000 patients and found the diagnosis rate of high-g...
The Data Science Institute team received the Facebook Systems for Machine Learning Award for their research on co-training machine learning models. The project aims to enhance machine learning models for scalability and adaptability, particularly in recommendation systems used by Netflix, Amazon, and Facebook.
Researchers at the Data Science Institute, Columbia, are designing intelligent headphones that detect sounds of approaching vehicles and send audio alerts to pedestrians. The system aims to counter growing public safety concerns caused by twalking, which has tripled in US injuries and deaths over seven years.
A new study from Columbia University suggests that planting less rice and more nutritious crops like finger millet, pearl millet, and sorghum can enhance India's food supply while reducing environmental impact. This diversification could increase protein by 1 to 5%, iron supply by 5 to 49% and reduce greenhouse gas emissions by 2 to 13%.
A Columbia professor has developed a machine-learning based detector that automatically detects and stops lateral phishing attacks within organizations. The detector uses features like anomaly detection in communication patterns to flag suspicious emails with high precision and low false positive rates.
Researchers at Columbia University have successfully controlled a visual behavior in mice by activating specific groups of neurons in their visual cortex. The study used high-resolution optogenetics and two-photon calcium imaging to identify and target individual neurons, demonstrating the causal role of neuronal ensembles in behavior.
The Northeast Big Data Innovation Hub has been awarded a $4 million grant from the NSF to build cross-sector partnerships, spur economic development, and accelerate big data innovation. The hub will focus on mission-driven projects that coordinate and stimulate translational data science.
The study found that alternative grains such as millet, sorghum, and maize are more resilient to extreme weather, while rice experiences larger declines in yields. This suggests that diversifying India's crop production can help adapt to climate change and improve nutrition.
Researchers at Columbia's Data Science Institute used cellular resolution imaging to investigate changes in neuronal microstates during anesthesia. They found that anesthesia disrupts local network dynamics, leading to breakdowns in macroscale connectivity.
A team of Columbia professors has designed a data-driven model to predict Li-Ion battery performance, aiming to reduce error rates from five percent to one percent. The model can help extend battery life and improve electric vehicles' efficiency by predicting charge levels and identifying weak cells.
A research collaboration between Columbia University and a genomicist has identified specific genetic pathways associated with antidepressant resistance. The team hopes to develop new treatments that can circumvent resistance in millions of people who do not respond to current medications.
The team aims to develop a sorting-machine prototype using pupae images and machine-learning algorithms to process tens of thousands of images daily. They also design a robot to sort the pupae based on the result of the algorithm, with the goal of reducing tsetse populations in sub-Saharan Africa.
A recent study published in BioScience analyzed international food trade data from 1986 to 2010 to assess its impact on equitable access to food. The researchers found that trade plays a significant role in distributing food more equitably across the planet.
A Columbia professor has developed a predictive computer platform that analyzes all tumor types and predicts effective treatment options. The platform, OncoTreat, is based on advanced data science techniques, including information theory and ray-tracing.
A new study reveals that short-term stress in surgeons can lead to a 66% increase in mistakes during operations, which can cause bleeding, torn tissue, or burns. The research, published in the British Journal of Surgery, could lead to protocols aiming to reduce acute stress on surgeons.
Researchers will use Azure and AI tools to analyze Hurricane Maria's impact on El Yunque National Forest. The 28,000-acre forest was severely damaged, with thousands of trees felled.
The Data Science Institute at Columbia has awarded grants to five research teams to use data science to solve societal problems in cancer research, medical science, transportation and technology. The teams will work together to transform several fields throughout the university.
A research team from Columbia University has developed a molecular taxonomy for hair disorders, which will help diagnose diseases affecting the hair follicles. The taxonomy was created by analyzing more than three million data points and identified nearly 5,000 biological terms shared by groups of hair genes.
Columbia University's SCRIPTS system uses AI to process documents in low-resource languages, providing summaries and translations. The project aims to enhance efficiency for intelligence analysts worldwide.
The Data Science Institute has developed a method to tap into available RF-spectrum channels using energy-efficient sensors. This will enable future communication systems to flexibly share the spectrum, reducing strain on the finite resource.
The Data Science Institute developed a novel statistical method to measure predictivity in big data analysis. The approach allows researchers to compare their predictions to a theoretical baseline, enhancing accuracy. The team will help the New York City Department of Transportation assess complex social problems using big data sets.
The Columbia team will use machine learning and advanced techniques to filter out extraneous data and aid in the detection of gravitational waves. With a $1 million NSF grant, they aim to contribute to LIGO's historic breakthroughs and advance understanding of the cosmos.
A recent study by Tal Danino at the Data Science Institute demonstrates that bacteria in pancreatic tumors degrade a chemotherapy drug, Gemcitabine. The study found that antibiotics were effective in killing these bacteria in over 70% of mice, leading to rapid tumor progression without treatment.
A Data Science Institute professor is leading a research team to develop an intelligent headphone system that detects sounds of approaching vehicles and sends audio alerts to pedestrians. The project aims to reduce pedestrian injuries and fatalities in crowded cities by providing a warning system for pedestrians wearing headphones.
A team of researchers from the Data Science Institute at Columbia are developing tools to flag aggressive social media posts from gangs and ISIS, aiming to prevent violent escalation. They will use natural language processing and machine learning techniques to analyze millions of tweets from gang members in Chicago and ISIS recruits.