Articles tagged with Big Data
Researchers develop crowd-assisted deep learning system to analyze disasters, integrating human intelligence with AI models for better results. The project aims to improve AI's interpretability and accuracy in disaster assessment applications.
A new study using big data to examine LBD in Singapore's transport gig economy aims to reveal what enables LBD and how it can be promoted. The researchers propose a novel framework to measure individual drivers' productivity and skills, including their ability to anticipate demands and competition.
The new method uses ink-jet printers and hyperspectral imaging to create hundreds of thousands of material combinations in a single trial run. A cobalt-tantalum-tin compound was discovered that exhibits tunable transparency and acts as a good catalyst for chemical reactions.
A new study from Northwestern University finds that exploring diverse styles before exploiting a narrow area can lead to a career's greatest hits. Dashun Wang and his team analyzed data from over 2,128 artists, including Jackson Pollock, and found a consistent association between the 'exploration-exploitation' pattern and hot streaks.
Researchers created a massive virtual universe, Uchuu, consisting of 2.1 trillion particles in a computational cube spanning 9.63 billion light-years. The simulation allows for the study of dark matter and large-scale structure on an unprecedented scale.
A University of Arizona-led study found that drought and seasonal fluctuations in rainfall are larger drivers of evolutionary diversity than warm temperatures. The research team created maps of evolutionary diversity across North, Central and South America, revealing that deserts have more plant species compared to forests due to drought.
A new study in the Journal of Marketing uses big data from over 100 million social media user engagements to derive marketing insights. The research captures latent relationships among thousands of brands and across many categories, revealing a highly precise market structure. This allows product managers to identify potential threats ...
A large-scale study of over 300,000 children in Australia found that those hospitalised with chronic illnesses were at a significantly higher risk of poor academic performance. The study, published in Archives of Disease in Childhood, highlights the need for additional support for these students.
Researchers analyzed social media connections of over 4 million users across 10 countries, finding common characteristics and behaviors, including follow ratios and profile length. These findings can help tailor data analysis to cultural differences, improving marketing and information sharing.
Scientists at CiTIUS have developed a new fast support vector classifier (FSVC) that significantly improves data classification using Machine Learning techniques. The FSVC is much faster and operates with less memory than traditional approaches, making it suitable for large-scale classification problems.
A big data study from UNSW Sydney found that Australian cancer patients kept up their pharmaceutical treatments during last year's COVID-19 lockdowns. The researchers attribute the good news to relatively low rates of COVID-19 infections in Australia, which minimally impacted cancer treatment patterns.
Immunology researchers from The University of Queensland have identified UMAP as a powerful tool for analyzing large patient datasets. This method performed significantly better than PCA in reducing the complexity of big data, enabling accurate patient stratification and clustering. The findings could lead to the adoption of targeted t...
A study by Texas A&M University suggests that current US laws do not align with the American public's preferences for using big data in public health. The public prefers big data to be used for common good over individual or self-serving interests.
A study by Bentley University explores how the elderly use smart speaker technology at home, revealing heterogeneous use patterns. The results show that mornings and afternoons are more active, music and news are most prevalent, and simple commands dominate interactions.
A team of researchers from IPK used large-scale data to develop predictive models for yield stability in hybrid varieties of wheat. By analyzing over 13,000 genotypes and 125,000 yield plots, they were able to double the accuracy of their predictions.
A student from UNIST proposed a ship-arrival time prediction model based on Artificial Intelligence and won the grand prize. The model aims to improve the efficiency of shipping and port logistics by accurately predicting vessel arrival times.
Researchers propose a novel architecture, Med-BDA, to analyze healthcare big data, enabling real-time predictions and better patient treatments. The new approach uses Apache Spark technology to tackle complex data analysis challenges.
The use of primary care big data in understanding COVID-19 pharmacoepidemiology can help inform patient care and policy decisions. By analyzing interactions between medications and COVID-19 outcomes, researchers can identify potential treatments to improve patient outcomes.
A University of Massachusetts Amherst study recommends guidelines for the ethical handling of opioid use disorder information stored in the Public Health Data Warehouse. The research highlights concerns about public trust and potential misuses of big data, and proposes safeguards to prioritize health equity.
Researchers develop AI algorithms to optimize flight networks for resilience against storms, reducing delays and improving safety. The project aims to create a flight planning software that can automatically react to storms and recover the system.
Researchers developed a groundbreaking model that defines new geographical scales from mobile tracking data, bringing geography back to understanding of mobility. The model identifies typical distances and choices corresponding to geographical boundaries, varying by individual characteristics.
Justin Zhan, a data science professor at the University of Arkansas, has received a $1.25 million grant to develop novel algorithms for enhancing computational speed and efficiency in applications requiring massive amounts of streaming data. His research aims to improve operational robustness, computational speed, and efficiency in too...
Researchers developed a deep-learning model that uses big data and artificial intelligence to predict future COVID-19 case growth. The model accounts for features such as mobility, population activities, and social demographics, achieving 64% accuracy in predicting cases.
Zhi Tian will receive funding to develop communication-efficient approaches for collaborative learning from private data in big data computing. Her goal is to minimize overall runtime, communication costs and total samples used.
New research highlights potential cardiovascular risk of novel anti-osteoporotic medicine romosozumab. The study found a link between genetic markers and increased cardiovascular risk, supporting previous trial findings.
A recent study explores the application of deep learning in ecological resource research, addressing challenges such as multi-source/multi-meta heterogeneity and high dimensional complexity. The study highlights the potential of deep learning in connecting computer science with classical theoretical sciences in ecology.
A new commentary paper highlights the urgent need to address environmental degradation using big data and technological advances. The study shows that despite increased computing speeds and data storage, the planet is still facing serious declines in forest cover and tidal flats.
A mobile contact tracing app can reduce transmission at any stage of the epidemic, helping to bring the pandemic under control. The app uses low-energy Bluetooth to log close proximity contacts and alerts users if they become infected, supporting health services and reducing serious social impacts.
A coronavirus mobile app could significantly help contain the spread of the virus, according to Oxford University experts. The team recommends deploying the app as part of an integrated control strategy that identifies infected people and their recent contacts using digital technology.
By using approximations instead of explicit kernels, researchers have accelerated machine learning speeds and improved AI's ability to handle large datasets. The new approach uses statistics to derive a nearly accurate kernel that can be computed much faster than traditional methods.
A new study by Michigan State University quantifies soil and landscape features and spatial and temporal yield variations in response to climate variability. The research identifies areas within individual fields where yield is unstable, with over one-quarter of corn and soybean cropland in the Midwest experiencing this issue.
Researchers using big data analytics have identified over three quarters of Spanish-founded colonial settlements in the former Inca Empire, providing new insights into social life and population history. The use of high-resolution satellite imagery also raises concerns about individual privacy and national security.
A new study highlights the need to balance research benefits with patient privacy concerns as big data is increasingly used in medical care. Researchers suggest improved education and legislation are necessary to protect consumers' sensitive health data.
RUDN University mathematicians developed a model to optimize data center efficiency using Markov chains. Their method reduces server overheating and improves server capacity utilization, resulting in significant cost savings.
The BC2 Basel Computational Biology Conference aims to bridge the gap between Big Data and clinical applications, featuring renowned experts from precision oncology to infectious diseases. Key takeaways include the importance of single-cell data in cancer research and biological big data analysis methods.
KAUST researchers develop a universal framework for querying big data, allowing researchers to focus on advancing the query engine rather than coding for specific platforms. The approach uses sparse-matrix algebra and achieves performance comparable to existing specialized approaches.
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.
Recent Chinese AI research has made significant breakthroughs in big data analysis, with a focus on developing algorithms to detect abnormal data regions and repair historical correlations. Researchers have also proposed using Doppler measurements to improve target tracking performance in noisy environments.
Researchers developed a series of models that strongly predict how terrain slope affects human travel rates, accounting for variability in movement. The study used crowdsourced fitness-tracking data from nearly 30,000 people, resulting in more advanced models than previous ones.
Researchers connected big data marketing tools to land conservation efforts, identifying landowners more likely to install riparian buffers. The study found that using microtargeting techniques could increase the impact of conservation programs and reduce outreach costs.
A team of international scientists used big data analysis to study how neurons communicate with each other in the brain, identifying patterns related to memory and discovering major proteins responsible for changes observed in neurons.
A study of over 8 million albums from 1956 to 2015 reveals that new musical styles emerge as a result of counter-signaling from outsider groups. This challenges traditional theories on the evolution of fashion and trends in music, highlighting the role of elite competition in driving innovation.
Developed with NSF funding, SETA (Scalable Event Trend Analytics) is an open-source software that analyzes high-volume data streams in real time to provide actionable insights. It helps organizations make data-driven decisions quickly, enabling applications like autonomous vehicle networks and healthcare.
Researchers urge scientists to utilize massive open-access data resources to address global biodiversity issues and forecast plant life's impact on a human-dominated planet. Big data can help make timely diagnoses and prescribe treatment plans for the planet.
A research team led by Heng Huang aims to create a framework for secure and efficient multi-site collaborative big brain data mining. The project addresses computational challenges in analyzing complex brain disorders and genomics data.
Researchers are using big data and machine learning techniques to optimize sports performance, game-day decision making, and even predict the end of daily fantasy sports. The special issue explores various applications of big data in sports analytics, from pacing strategies in long-distance running to business-side insights.
A new study by MIT researchers finds that compiling massive, anonymized datasets about people's movement patterns can make it easier to discern users' identities. The study shows how merging different types of location-stamped data can lead to a high matchability success rate, increasing the possibility of deanonymizing real user data.
Archaeologists used machine learning techniques to classify and predict the technological elements of ancient hunter-gatherer groups in Patagonia. The study identified two distinct 'landscapes' of technology, one for pedestrian groups and another for nautical societies, shedding light on their mobility patterns and interactions.
A RIT researcher is developing new signal processing solutions to improve data analysis efficiency and reliability. The project aims to reduce the impact of faulty measurements in complex sensing systems by creating algorithms that can detect and mitigate corrupted data.
By eliminating redundant data, researchers have developed a technique that reduces the amount of information needed for accurate predictions. This approach has been successfully applied to various applications, including soil quality prediction, healthcare, and environmental studies.
The University of Pittsburgh has received a $1.2 million NSF grant to analyze electronic anesthesia records and prevent postoperative complications and death using machine learning and Big Data analysis. Dr. Heng Huang will develop a new deep learning algorithm to predict surgical outcomes based on historical patient data.
Researchers at Cold Spring Harbor Laboratory have developed a new approach called Density Estimation using Field Theory (DEFT) to analyze small datasets, inspired by theoretical physics. The method fixes shortcomings of common statistical methods, providing more certainty in conclusions.
The article discusses the complexity of data ethics, emphasizing personal data ownership, consent, trustworthiness, and privacy. Researchers aim to strike a balance between harnessing data's potential benefits and mitigating its risks.
A synergistic approach to big data in science emphasizes the importance of collaborative research efforts and open science. Ecologists must work together to answer complex, globally relevant questions that cross disciplines and require extrapolating findings from one location to another.
The article highlights the need for harnessing technology to analyze healthcare data and generate new evidence, which can be combined with published reviews to improve health outcomes. Michigan Medicine's Knowledge Grid platform is taking the lead in transforming biomedical knowledge into computable forms that can inform medical practice.
The review identified gaps in using technologies, with a lack of information on extreme temperatures and flooding. Despite limitations, big data and ICT hold promise for potential solutions to harness diverse and chaotic data in disasters.
The ZPID Twin Conference brought together over 150 participants to discuss Big Data in Psychology and Research Synthesis. Keynote speakers highlighted the importance of psychology's theoretical strength in addressing big data challenges, while also emphasizing the need for innovative methods and tools for research synthesis.
PlinyCompute is a system designed for developing high-performance big data codes, offering faster implementation of complex object manipulation and library-style computations compared to Spark. The platform was developed by Rice University's DARPA-funded Pliny Project team, which aims to create sophisticated programming tools using mac...
The consulting firm is founded by four computer science experts from Saarland University to provide sound advice on data analysis. Data Science Consulting focuses on collecting, cleaning and merging data, as well as removing errors, maintaining data, setting up a scalable architecture and defining critical characteristics for analysis.
The FraudBuster approach uses proactive risk prediction at the underwriting stage to identify unprofitable drivers who are likely fraudulent risks. This novel method can help insurers reduce fraud in high-risk markets, such as the automobile insurance market.