Articles tagged with Data Sets
Researchers developed an AI tool called SILIC to identify 169 species, including 137 birds, from bird sounds in Yushan National Park. The dataset provides detailed acoustic activity patterns of wildlife across short and long temporal scales.
A machine-learning model was trained on 10 million tweets to infer users' subjective wellbeing. The study found that New Year's Day is the most popular holiday, followed by Defender of the Fatherland Day and International Women's Day. The researchers also discovered gender differences in attitudes towards certain holidays.
Researchers developed AI models based on UNet and MobileNet architectures to analyze standardized abnormalities in CT images, accurately identifying object presence and confidence. These models achieved an absolute percentage error of less than 5 percent, comparable to human professionals.
Researchers found that COVID-19 infections are linked to an increased risk of developing liver problems, acute pancreatitis, and other GI disorders. The study analyzed over 14 million medical records and estimated that SARS-CoV-2 infections have contributed to over 6 million new cases of GI disorders in the US.
A novel AI architecture, relational reasoning network, accurately identifies anatomical landmarks in CT scans for orthodontic treatments. The model learns spatial relationships between landmarks without explicit image segmentation, achieving accuracy comparable to conventional methods.
Researchers seek to develop algorithms providing meaningful explanations for AI decision-making, enabling higher human trust and adoption in fields like science. The project focuses on symbolic reasoning and estimating explanation accuracy, addressing the need for transparent AI systems.
Researchers at MIT developed a technique to improve machine-learning models' reliability without requiring additional data or extensive computing resources. The method uses a simpler companion model to estimate uncertainty, enabling more effective uncertainty quantification.
Researchers tested three common techniques to make algorithms fairer and found that one approach didn't reduce social norm bias at all. They proposed a new technique: a formula to directly measure social norm bias in an algorithm so it can be corrected. This bias can persist even after overt discrimination is removed.
Researchers found that electric car adoption in California was associated with real-world reductions in air pollution and asthma-related emergency room visits. The study also highlighted an 'adoption gap' between low-resource zip codes, pointing to opportunities for environmental justice.
Dr. Nico Spiller to develop new analysis methods using machine learning to analyze complex brain data related to memory, decision making, and movement. The fellowship aims to provide insights into neurodegenerative diseases such as Alzheimer's and Parkinson's disease.
Researchers identified 7 key symptoms of long COVID, including heart issues and joint pain, in a study of 52,461 patients. The findings could help healthcare providers diagnose and treat the condition more effectively.
A new measure called c-value helps researchers choose between techniques based on the chance that a new method is more accurate for a specific dataset. The tool answers questions like whether to use alternative estimation methods despite potential costs and effort.
A new study used satellite data and public registry information to track the changing identities of commercial fishing vessels, revealing that nearly 20% of high seas fishing is carried out by unregulated or unauthorized vessels. The study found hotspots of potential IUU fishing in the Southwest Atlantic Ocean and western Indian Ocean.
A recent study published in Nature has discovered several new disease genes and provided new insights into the effects of known genetic factors on disease. The study highlights an underappreciated complexity in dosage effects of genetic variants, challenging traditional Mendelian inheritance laws.
A team of scientists reviewed the effectiveness of reanalysis data products for studying the West African climate. They found that ERA5 achieved considerable progress in reducing biases and improving representation compared to its predecessor, ERA-interim.
Researchers identified a common brain network underlying several psychiatric illnesses, including schizophrenia and depression. The transdiagnostic network shows gray matter decreases in specific brain regions across most studies.
Researchers at Drexel University used GPT-3 to spot early signs of Alzheimer's in spontaneous speech, achieving 80% accuracy. The program analyzed word-use, sentence structure and meaning from transcripts to identify characteristic profiles of Alzheimer's speech.
A recent study has identified common and unique cellular processes in six neurodegenerative diseases, providing new insights into the underlying causes of these conditions. The research used machine learning analysis to compare RNA markers in whole blood samples from patients with distinct diseases, revealing eight shared themes across...
A team of ecologists is calling for the designation of World Heritage Environmental Datasets to secure funding and ensure their long-term accessibility. These datasets, which include vital information on climate change adaptation, resource management, and environmental policy, are essential to understanding global change.
Researchers analyzed data from over 80,000 primary care physicians and found no clear connection between MIPS scores and clinical performance. Doctors with low MIPS scores performed worse on some process measures but better on others, suggesting that the system may prioritize paperwork over patient outcomes.
Scientists used AI-driven PandaOmics platform to analyze gene expression datasets from DNA repair diseases, identifying biomarkers associated with treatment response. The study focused on genes that stratify cancer patients by survival outcomes, providing potential targets for personalized therapies.
The Washington D.C. metro area's dashboard is being developed at PSU, allowing users to see all the data together in one place. The project aims to improve nonmotorized planning by providing clean, quality-checked biking and walking count data.
A new study maps the global landscape of antimicrobial resistance, revealing surprising transmissions in Sub-Saharan Africa and highlighting the need for tailored strategies to combat resistance. The research, which analyzed sewage samples from 243 cities in 101 countries, found that resistance genes are more frequently transmitted acr...
Researchers have discovered that oligodendrocyte precursor cells (OPCs) play a crucial role in synaptic pruning, cleaning up unwanted connections between neurons. By analyzing a massive dataset of 3D brain cell structures, the team found OPCs digesting parts of neighboring neurons.
A global dataset reveals widespread witchcraft beliefs varying substantially between countries and world regions. Higher education and economic security are associated with lower belief in witchcraft. Weak institutions, low social trust, and conformist culture also correlate with higher witchcraft prevalence.
Chung-Ang University researchers propose a new algorithm, MR-UCB and MR-APE, to tackle stochastic multi-armed bandit problems with heavy-tailed noise distributions. The methods guarantee minimal loss for worst-case scenarios with minimal prior information.
Researchers created synthetic knee x-ray images to complement real images in osteoarthritis classification. Medical experts were unable to distinguish between authentic and synthetic images, highlighting the potential of synthetic data for collaboration and testing.
The Julich Brain Atlas provides detailed maps of brain cells and receptors, enabling better understanding of brain connectivity and function. The atlas allows for correlation between brain network activity and underlying anatomy, aiding in the diagnosis of psychiatric disorders.
Researchers modelled relationship between plant diversity and environmental conditions, capturing how diversity varies along environmental gradients. The models predict highest concentrations of plant diversity in environmentally heterogeneous tropical areas like Central America and the Amazonia.
Researchers at KAUST develop a novel multivariate skew-elliptical link model to address the challenges of highly imbalanced health data. The new model provides a better fit to COVID-19 datasets and offers flexibility over existing models.
Researchers at University of Jyväskylä used machine learning to predict ACL injuries in elite athletes but found a low overall accuracy rate. The study analyzed the largest data set ever collected and provided valuable insights into the challenges of predicting injuries in individual athletes.
A new study by MIT researchers shows that mobile phones can collect useful structural integrity data while crossing bridges. The study found that information about bridge vibrations can be extracted from smartphone-collected accelerometer data, and that this method could add years to a road bridge's lifespan. By leveraging crowdsourced...
The Global Jukebox, an online tool for exploring music and performing arts from around the world, has made its dataset and data available to the public. The database includes 5,776 recordings representing 1,026 societies, with detailed musical style categorization data and additional features such as breath management and instrumentation.
Researchers used machine learning to track turbulent structures in fusion reactors, gaining detailed information on their behavior and heat flows. The approach enables more accurate engineering requirements for reactor walls and could lead to improved energy efficiency.
The US Department of Energy's Oak Ridge National Laboratory has developed a massive geographic dataset, USA Structures, using deep learning to forecast potential damage and accelerate emergency response. The dataset provides critical information on building outlines and attributes, enabling FEMA to prioritize response efforts.
A joint study by TAU and Hebrew University accurately dated 21 destruction layers at 17 archaeological sites in Israel, using geomagnetic field reconstruction. The new data verify Biblical accounts of Egyptian, Aramean, Assyrian, and Babylonian military campaigns against the Kingdoms of Israel and Judah.
Researchers analyzed COVID-19 conspiracy theories on Twitter, finding that some peaked at the beginning of the pandemic while others remained persistent throughout. The study suggests that risk perceptions and lack of information from governments and experts contributed to the persistence of these theories.
A new dataset provides an unprecedented insight into the Arctic Ocean's biological life, revealing a year's worth of microbial communities and their responses to climate change. The EcoOmics dataset aims to guide conservation efforts and provide evidence for novel biology that may influence our understanding of evolution on Earth.
A study found that Italian regions with high levels of environmental pollution have higher cancer mortality rates, even after controlling for lifestyle factors. The analysis identified specific sources of pollution associated with certain types of cancer, highlighting the need for environmental reduction and prevention to combat cancer.
A University of Groningen team created two machine learning models to predict app removal risks, achieving accuracy rates of up to 79.2%. The models can help developers avoid bans and users protect their data.
The Keck School of Medicine's Stevens INI has received a nearly $2 million grant to upgrade its storage system, increasing its capacity to handle vast amounts of brain imaging data. This will enable the lab to maximize scientific productivity and fulfill its promise to collaborate with researchers globally.
A new study by Cornell University researchers found that nearly all market research panel participants are at risk of being de-anonymized due to the presence of quasi-identifiers. The reidentification risk can be up to 94% when considering multiple observations per panelist, highlighting the need for improved data privacy measures.
Researchers aim to create a unified database network for battery data, facilitating AI analysis and predictions. The Battery Data Genome will collect data across the entire battery lifecycle, from discovery to deployment, with uniform standards for metadata.
Researchers developed a tool that encodes patient data as DNA sequences to link health databases accurately. The platform uses BLAST and machine learning algorithms to integrate data from multiple administrative databases, overcoming typographical errors and inconsistencies.
Researchers found that 12% of cancer survivors lived in poverty, experiencing negative effects on physical and mental health. Financial burden of cancer care persists years after diagnosis, leading to poor health outcomes and increased mortality.
Researchers found that a common autism screening test consistently excludes more women than men from research studies, creating a 'leaky pipeline' for diagnosis and treatment. This bias is attributed to the test's origins in male-dominated samples, which may not accurately capture female phenotypes.
Researchers from Children's Hospital of Philadelphia used advanced mapping techniques to identify causal genes and target pairings in the pancreas linked to type 2 diabetes. The study revealed alpha and acinar cells play a greater role in disease development than previously thought.
Researchers at King Abdullah University of Science & Technology (KAUST) have developed a more robust and realistic general method for dealing with wind-driven phenomena in geostatistical modeling, which promises to greatly improve the accuracy of pollutant dispersion prediction.
A new study from North Carolina State University shows that soil temperature can be used to predict the spread of the corn earworm, an important pest affecting various crop species. The research reveals three geographic zones where the pest can overwinter, and models suggest that these zones will shift northward due to climate change.
Rice University's ROBE Array algorithm slashes the size of DLRM memory structures, allowing training on 100 megabytes of memory and a single GPU. The method matches state-of-the-art DLRM training methods with improved inference efficiency.
Thirty-eight Chinese cities have reduced their CO2 emissions for at least five years, while 21 cities have cut emissions due to economic decline or population loss. The study recommends individualized emission targets considering cities' resources and development goals.
Mandated PDMP use in US states reduced opioid prescriptions by 6.1% but led to a 50.1% increase in heroin-related deaths, according to researchers from the University of Texas at Dallas.
Researchers developed a digital health system approach to enhance clinical decision-making for low-back pain sufferers. The system combines self-reported pain measures with precise motion-sensing data, revealing functional improvements only after six months of surgery.
A new Harvard-led study reveals that rising global temperatures caused the explosive evolution of early reptiles, challenging previous explanations. The research suggests climate change triggered morphological changes in reptile groups, including those that gave rise to crocodiles and dinosaurs.
Researchers assembled a large biologging dataset to understand how sharks, rays, and skates use the ocean depths. They found that many species migrate vertically in response to food resources, with some diving to extreme depths.
A new study from North Carolina State University reveals no-till farming boosts agricultural land values in the Midwest, with a 1% increase translating to a $7.86 per acre gain. This finding supports the economic and environmental benefits of adopting soil conservation practices like no-till farming.
A new study found that 'gaps' within DNA protein and sequence alignments can provide valuable information about nucleotide and amino acid substitutions. The finding is particularly relevant for studying distantly related species.
Researchers at Cedars-Sinai created complex computer models of individual brain cells using artificial intelligence. These models capture the electrical signals that neurons fire to communicate with each other, allowing researchers to replicate brain activity at the single-cell level.
Researchers aim to uncover information on blotchy bass syndrome's prevalence, distribution, and potential risks through data collected by citizens. The study will help determine which bass species are susceptible and which seem immune.
The new database offers improved estimations of Potential Evapotranspiration and Aridity Index for the entire world at a high spatial resolution. This allows for a finer description of water needs in different regions, enabling better resource management and land-use planning.