Articles tagged with Big Data
A UTSA computer science professor has received a Department of Army grant to create an image searching algorithm for combing through large amounts of surveillance videos. The goal is to identify individuals more quickly, especially in crowded and hectic situations like the Boston Marathon bombing.
The BigStorage project aims to develop new approaches for handling Big Data, leveraging theoretical research, complex infrastructures, and software packages. The European consortium will focus on high-performance computing and storage technologies to address the needs of climate research, medicine, and environmental sciences.
The Alan Turing Institute has made significant progress in its first few days of operations, with the appointment of Professor Andrew Blake as its first Director. The Institute will promote the development and use of advanced mathematics, computer science, algorithms, and big data for human benefit.
Scientists from InSilico Medicine have developed an approach to screen and rank geroprotective drugs using big data analysis, identifying compounds with potential geroprotective properties. The Geroscope software was applied to gene expression data derived from stem cells to select five drugs that displayed geroprotective action.
Researchers at Case Western Reserve University are developing a platform to collect, analyze disparate clinical information from multiple sources, ensuring comparability and reproducibility. The goal is to integrate datasets for enhanced approaches to care and treatment of conditions like epilepsy and lung cancer.
The Person-Event Data Environment (PDE) database captures financial, health, medical and other data for every soldier from entry to separation. The database includes information on soldiers' military experience, contractors and dependents, representing a heterogeneous population with various socioeconomic backgrounds.
The five-year research project will train underrepresented minority students in STEM fields to address a critical shortfall in the workforce essential for future NASA missions. The project will include undergraduate training, research, and doctoral studies, aiming to develop new approaches to visualize and analyze massive data sets.
The US National Science Foundation and Japan Science and Technology Agency announce joint support for six projects to improve future disaster management. Researchers will leverage Big Data and data analytics to capture and process disaster data, enhance resilient networks, and develop novel approaches to analyze large, noisy data.
Researchers used big data from HapMap and 1000 Genomes projects to discover the gephyrin gene's connection to human history and its role in complex neurological diseases. They found a region with rapid evolution after splitting into yin and yang haplotypes, prevalent across different populations.
MSU researchers are using big data analytics to help farmers adapt to climate variability, reducing nitrous oxide emissions and algal blooms. The project aims to integrate crop models with satellite imagery and UAVs to promote water-, nutrient- and climate-smart technologies.
Columbia engineers invent nanoscale IC that enables simultaneous transmission and reception at the same frequency in a wireless radio, doubling data capacity. The technology, known as full-duplex radio integrated circuits (ICs), cancels transmitter echo and enables conversations to take half the amount of time.
Academics from the University of Bristol's Intelligent Systems Laboratory used big data to analyze mass media coverage of the 2012 US presidential election. The study found that the media focused more frequently on positive statements about the Democrats, while the Republicans were often portrayed in a negative light. Key issues covere...
A new study by University of Vermont researchers confirms that humans use more positive words than negative ones across ten languages, including Arabic, Korean, and Chinese. The study analyzed billions of words from various sources and found a consistent positivity bias, indicating that language itself has a positive outlook.
The use of big data to detect disease outbreaks has the potential to strengthen global public health surveillance, but raises ethical questions about privacy, consent, and legitimacy. The authors propose a framework to address these challenges, including adapting requirements for public health contexts and ensuring methodological robus...
UC San Diego researchers use social network analysis to combine Google Flu Trends' 'big data' with CDC data, improving predictions. The study refines Google's predictions by weighting them with a social network derived from CDC reports on laboratory-tested cases of flu.
A new algorithm, TopicMapping, has been developed to improve the accuracy and reproducibility of big data text analysis. By using a network approach, the algorithm separates unstructured text into topics with high accuracy and reproducibility. The results show that existing algorithms, such as LDA, are not reliable for complex datasets.
The EMBERS system uses publically available data to predict population-level societal events like civil unrest and disease outbreaks. The system processes large volumes of social media data and makes predictions using various modeling approaches.
The UTSA Cloud and Big Data Laboratory is partnering with Indiana University on a $6.6 million NSF grant to create a research cloud for thousands of researchers, providing easy access to advanced computing tools. The project, called Jetstream, will utilize OpenStack-based cloud environment.
A new study suggests that online searches for specific car features can predict consumer purchases with greater accuracy than brand names. Researchers analyzed six years of car sales data and found that searches for features like body type and fuel economy were key indicators of future purchases.
Researchers from the Keck School of Medicine of USC led a global consortium to identify eight common genetic mutations that appear to age the brain an average of three years. The discovery could lead to targeted therapies and interventions for Alzheimer's disease, autism, and other neurological conditions.
The Biogerontology Research Foundation will present new economic longevity research at the second Big Data Science in Medicine congress in Oxford. The research, recently published in Psychology Research and Behavior Management, details an extensive survey of International Employee Benefits Association members.
A study of over 9,000 urban data sets from 20 cities reveals the growing availability and quality of these datasets, enabling better analysis and decision-making. The findings highlight the importance of integrating different data sets and overcoming challenges in accessing and utilizing this valuable resource.
A study by Northeastern University researchers found evidence of price discrimination on four general retailers and five travel sites, with significant price differences in some cases. The team examined 16 popular e-commerce sites using a sophisticated methodology to identify price steering and discrimination.
Researchers explore challenges and opportunities of mining large climate datasets with powerful analytical methods, scientific theory, and solid data engineering. By combining theory and Big Data, scientists aim to explain and predict important climate change phenomena.
A $9.2 million grant from the NIH will help researchers share, use and cite biomedical datasets more efficiently. The project, led by UC San Diego, aims to create a searchable online digital library for health-related datasets.
Researchers will develop the KnowEnG tool, integrating multiple analytical methods for intuitive genome-wide data analysis. The center aims to create a powerful computational tool that offers new functional insights for genes being studied.
The Scripps Research Institute and Scripps Translational Science Institute will receive funding for a new Center for Excellence in Big Data Computing, enabling the collective mining of biomedical data. The initiative aims to accelerate research efforts on rare diseases using open approaches and citizen science platforms.
The Center for Big Data in Translational Genomics aims to develop standard protocols and tools for handling genomic data efficiently, enabling the analysis of millions of genomic datasets. The center will test new approaches in four pilot projects, including cancer-related initiatives.
The University of California, Los Angeles (UCLA) will develop new strategies for mining and understanding complex biomedical data sets. The center's findings will help shape guidelines for future data integration and use.
A Big Data to Knowledge Program grant will develop big data methods to tackle 80% of coronary heart disease causes, including physical inactivity and poor stress management. The goal is to create adaptive, individualized health-behavior interventions using real-time data from smartphones and wearable devices.
The University of Pittsburgh has been awarded a four-year, $11 million grant to lead a Big Data to Knowledge Center of Excellence. The center will develop and disseminate tools that can find causal links in large biomedical data, enabling researchers to uncover new insights in health and disease.
Daphne Yao aims to provide trustworthy data and infrastructure for military networks, detecting abnormal action sequences and workflows. Her research has shown promising results, enabling a leap forward in Army command and control of cyberspace capabilities.
The era of Big Data presents new opportunities for personalized treatments and services, but also introduces challenges in statistical analysis, bias sampling, and measurement errors. Researchers focus on developing new computational infrastructure and data-storage methods to handle massive datasets.
Researchers are leveraging parallel computing in the cloud to accelerate discovery and innovation in personalized medicine. By harnessing supercomputer powers, scientists can tackle previously infeasible problems and make groundbreaking discoveries.
Researchers developed XRay, a first step in understanding online personal data transparency. The tool analyzes data inputs and outputs to detect abusive practices, revealing sensitive topics and obscure targeting behaviors.
Researchers successfully ran 10,240 parallel simulations of global weather using Japan's K computer, achieving a significant improvement in forecasting accuracy. The study found that faraway observations can have an immediate impact on weather forecasts, highlighting the need for advanced methods to utilize this data.
A UT Arlington big data team has won a three-year $600,000 NSF grant to develop an interactive database of fruit fly gene expressions, which are commonly found in humans and other species. The project aims to yield methods of analyzing data that will aid in biomedical science and engineering.
Analyzing six use cases for big data in healthcare, researchers found strong opportunities for cost savings in high-cost patients, readmissions, and treatment optimization. However, bed shortages were associated with higher mortality rates, highlighting the need for a public health approach to address this issue.
A new study outlines opportunities to reduce healthcare costs through the use of big data. Researchers highlight five key areas: high-cost patients, readmissions, triage, decompensation, and adverse events. By analyzing large datasets, clinicians can identify high-risk patients, predict readmissions, and optimize treatment for chronic ...
Researchers at Duke University Medical Center argue that new technologies, including electronic health records and wearable monitoring devices, enable patients to provide real-time information on symptoms and clinical outcomes. This shift towards patient-driven data collection enhances the accuracy and timeliness of health research.
The big data era presents both challenges and opportunities for modeling and simulation, as it requires adapting to novel research thinking and methods. The paper highlights the need for exploratory research combining simulation-based engineering and science to develop a new simulation paradigm.
The National Institutes of Health has awarded $600,000 to the University of California, Riverside to expand its data storage capacity. This grant will enable researchers to process vast amounts of high-throughput data generated in various fields, including genome biology and biomedical sciences.
The emergence of big data presents both challenges and opportunities for databases, necessitating the development of new technologies and systems to handle vast amounts of complex data. Various distributed processing frameworks and systems have been proposed to address these issues.
A study by University of Houston researchers found that Google Flu Trend overestimated flu cases in the US by up to 50% and failed to accurately predict levels in previous seasons. The researchers suggest combining big data with traditional methodologies for a more accurate understanding of human behavior.
Researchers at Cold Spring Harbor Laboratory have proposed a new method for analyzing Big Data using mutual information, which can reveal patterns without prior assumptions. This approach challenges the latest statistical tools and has the potential to greatly benefit modern data analysis in biology and other fields.
Researchers found that certain behaviors may not be observable without massive data, highlighting the importance of considering data quantity in predictive modeling. The study suggests that sweeping assumptions about bigger being better can be dangerous and that the power of analytic tools depends on their appropriate use.
A new risk measure called a home score combines patients' symptoms and demographic information with local strep throat activity to estimate an individual's strep risk, empowering them to seek care appropriately. This innovative tool has the potential to eliminate 230,000 unnecessary doctor visits for strep throat in the US annually.
The winners of the 2013 Semantic Web Challenge have successfully combined innovative semantic web technology with an end user practical focus and efficiency in use. This year's challenge saw 17 proposals, with four Open Track Challenge winners and one Big Data Track winner selected by a panel of experts.
A new collaboration describes a new centipede species using a holistic approach combining 3D imaging, DNA barcoding, transcriptomic profiles, and video of the living specimen. The 'cyber-type' allows for global access to the specimen's data, enabling faster conservation efforts.
A new approach to describing new species uses big data techniques like next-gen sequencing and barcoding. This allows scientists to create detailed datasets and make it easier to manage biodiversity information.
A UT Arlington computer scientist is leading a project to mine electronic medical records data to personalize patient treatment, predict health care needs, and identify risks. The goal is to improve healthcare outcomes by providing doctors with better information.
LSU researchers have received a nearly $1 million grant to develop software and infrastructure enabling Big Data research on the university's campus. The project, titled CC-NIE Integration, will empower scientific breakthroughs by providing advanced information technologies and cyberinfrastructure.
A team of researchers has developed a program called RealBrush that allows graphic artists to quickly produce realistic brushstrokes on their computers. The program uses machine-learning approaches and Big Data storage techniques to create, bend, and shape various types of brushstrokes.
Rice University researchers are developing a customized, energy-efficient optical network called BOLD to handle the growing amounts of data in various scientific fields. The new network will utilize optical switches with high capacity and low power consumption, enabling faster processing and analysis of large datasets.
The Collaborative Assessment and Recommendation Engine (CARE) system uses electronic medical records to offer rapid advances in personalized health care, disease management, and wellness. CARE generates personalized disease risk profiles based on collaborative filtering method and Big Data science.
The study examined how Lumosity's dataset can provide insights into the lifestyle correlates of cognitive performance. The second study found that performance on tasks that rely on fluid intelligence decreases with age at a faster rate than those that rely on crystallized intelligence.
The MacArthur Foundation has awarded Carnegie Mellon University a $175,000 grant to investigate the role of social media and big data analytics in advancing human rights protection. The project aims to analyze the effectiveness of these technologies in gathering and analyzing data on human suffering and political repression.
A team of researchers at Saarland University developed the Hadoop Aggressive Indexing Library (HAIL), a technique that enables fast and efficient searching in large datasets. By generating indexes for multiple criteria, HAIL can query big datasets up to 100 times faster than traditional methods.
The inaugural issue of Big Data journal launched at the Strata Conference, featuring premium content from leading experts on big data. The Journal aims to facilitate dialogue among researchers, analysts, and policymakers on the challenges and opportunities in big data, published under the Creative Commons Attribution license.
Scientists have developed an automated system that can rapidly reconstruct hundreds of ancestral languages, including Proto-Indo-European and Proto-Afroasiatic. The computer program uses probabilistic reasoning and machine learning to replicate linguistic changes over time, with an accuracy rate of 85%.