Articles tagged with Data Mining
AI language models can extract and integrate information from vast amounts of unstructured environmental data, identifying pollutants and their toxic effects. While still in its early stages, the application of LLMs in aquatic risk assessment has the potential to support more dynamic and data-driven risk management strategies.
A new study using AI-powered BioBERT model accurately identifies fungal lifestyles, switching between helpful partner for plants to aggressive decomposers. The tool has nearly 90% accuracy and can scan thousands of papers in minutes, flagging species that may switch roles.
Jundong Li, an associate professor at the University of Virginia, has received the 2025 Tao Li Award for his significant contributions to data mining and machine learning. His research focuses on developing models that can extract actionable insights from structured data, particularly graphs.
A research group has constructed a 'material network' for metallic glasses, revealing that fully connected groups of elements are excellent predictors for new alloys. The study also found that many recent discoveries were already encoded in the network structure from earlier data.
Researchers from The University of Osaka developed MicroAdapt, a groundbreaking self-evolving edge AI technology that enables real-time learning and forecasting capabilities within compact devices. It achieves up to 100,000 times faster processing and 60% higher accuracy compared to state-of-the-art deep learning methods.
Researchers analyzed over 50,000 barrage comments from a promotional video on Chinese platform Bilibili, finding a clear link between positive emotional comments and purchasing intent. The study also observed real-time emotional contagion and nuanced dynamics in viewer behavior.
Researchers found cotton leafroll dwarf virus (CLRDV) infecting plants in southern US states as early as 2006, contradicting the assumption that it emerged more recently. The study used modern data-mining tools to uncover hidden threats and highlights the importance of maintaining accessible databases for disease surveillance.
University of Missouri researchers create digital sentiment map using AI to analyze public Instagram posts, linking emotional tone to real-life features. The tool aims to improve city services, identify areas of concern, and inform emergency response decisions.
Researchers developed a data-driven AI framework that identifies potential solid-state electrolyte candidates and predicts their performance. The framework uses large language models, multiple linear regression, and genetic algorithm to optimize battery design.
Dr. Latifur Khan, a renowned computer science professor, has been elected as an AAAS Fellow for his pioneering work in machine learning and big-data analytics. He developed innovative solutions to adapt machine learning models to cybersecurity risks and created an AI-driven tool to analyze political conflict and violence.
A report by University of Cambridge experts highlights the risks of unregulated AI use on UK's thriving creative sector. The researchers argue that forcing creatives to 'opt out' of AI training could compromise their livelihoods, particularly those just starting out.
The two-year study aims to explore biases in AI systems and develop a 'human-in-the-loop' framework for quality data discovery. It will investigate how humans can be involved as labelers, prompters, and validators to improve data sets and user interfaces.
Scientists at Anglia Ruskin University are collaborating with Medannex to accelerate treatment for bone cancer in children, focusing on paediatric osteosarcoma. The first-in-class therapy MDX-124 has shown promising results in preclinical tests and is being evaluated in a clinical study.
A new study in Health Data Science leverages keywords from Google Trends alongside research abstracts from the WHO COVID-19 database to dissect the pandemic's discourse dynamics. Academics respond faster and provide more detailed insights than GT, enriching policy formulation.
Researchers at UC Santa Cruz find that popular link prediction metrics are flawed and do not accurately measure algorithm performance. They recommend using a new metric, VCMPR, to benchmark link prediction tasks and highlight the importance of accurate metrics in machine learning decision-making.
Researchers at Tohoku University have developed an integrated approach to discovering stable and low-cost electrocatalysts, using data mining to accelerate the transition to renewable energy. The study identified 68 promising metal oxide electrocatalysts under specific conditions, including Sb2WO6 for oxygen reduction.
Scientists at Max-Planck-Institut für Eisenforschung developed a machine learning model that enhances predictive accuracy in alloy design, uncovering new corrosion-resistant compositions. The model combines numerical and textual data, enabling the identification of optimal alloy formulas.
The LY6 gene family has been found to be overexpressed in uterine corpus endometrial carcinoma (UCEC), leading to poor patient survival. Several LY6 genes have been identified as potential tumor-associated antigens and biomarkers for UCEC detection and prognosis.
A study by Drexel University and Vanderbilt University analyzed 82 relevant conversations on Instagram direct messages where teens asked for help, revealing that most disclosures were about mental health concerns. Support was offered in most cases, but specific sets of circumstances led to denial.
Researchers found that people dislike fitness recommendations from apps using social media data, instead favoring personalized approaches with control options. Users prefer systems allowing them to choose between different recommendation methods, making them feel more in control.
Researchers from George Mason University found that anti-tobacco campaigns on Facebook discussing the risks of second-hand smoking to pets received the most user engagement. The studies analyzed factors influencing effective antismoking campaigns and user interaction.
A group of researchers from Chung-Ang University have summarized the applications based on anomaly detection in multivariate time series. They evaluated the current state-of-the-art anomaly detection techniques and addressed the challenges associated with them, providing a thorough overview of the applications for anomaly detection in ...
A Washington State University-led research team created a product allocation model that leverages past customer transactions to provide brick-and-mortar stores with personalized shopping experiences. The model identifies profitable products, item associations, and past-aisle impulse to determine ideal product placement and optimize sto...
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.
Researchers provide a comprehensive catalog of medical knowledge graphs, their creation, and usage. Medical knowledge graphs have wide-ranging utility in the medical field, including disease diagnosis and drug development.
A new project aims to examine the circulation of newspaper reports on anti-Black violence between 1863 and 1921. The team will use computational methods to trace how stories spread across the country and map their impact, with potential applications for studying other forms of racial violence.
Engineers and chemists at the University of Illinois have combined electron microscopy and data mining to visualize chemical and physical alteration within ion batteries. The study reveals patterns of nucleation, growth, and coalescence that can inform the development of better rechargeable battery performance.
Researchers are developing a comprehensive surveillance system to track lingering symptoms of COVID-19, known as long COVID-19. They will utilize existing health data and enroll patients to gain accurate information on the burden and outcomes of individuals affected by long COVID.
Researchers at Washington University in St. Louis are evaluating the potential of AI to improve health outcomes and doctor well-being. Chenyang Lu's team has developed novel methods using deep learning to predict physician burnout and surgical outcomes, transforming clinical data into accurate predictions.
The BiCIKL Project has successfully integrated FAIR data on biodiversity for a year, focusing on creating a more transparent and efficient research ecosystem. The project aims to link different sources of biodiversity data across various infrastructures and open data platforms.
Researcher Jundong Li aims to improve machine recommendations and predictions based on cause and effect. He develops a suite of algorithms and mathematical models informed by human experience and intuition to find cause-and-effect relationships in big data.
A recent study by the University of Rochester found that mobility patterns can be predicted with surprising accuracy based on data collected from acquaintances, even if individual users turn off their own location tracking. The researchers discovered that up to 95% of an individual's movement pattern can be inferred from people they ar...
Researchers used a contrast pattern mining algorithm on publicly available data from 16,000 participants in the T1D Exchange Clinic Registry. The study found individuals with an immediate family history of Type 1 diabetes were more frequently diagnosed with hypertension and other co-occurring conditions.
Researchers at Rice University developed a new program called Variabel to accurately identify 'low-frequency' variants of the virus that causes COVID-19. By distinguishing true variants from sequencing errors, Variabel enables rapid characterization of within-host variation, which could aid in discovering future mutations.
A German-American research team predicts twenty-eight novel 2D materials with remarkable electronic and magnetic properties. The study utilizes a vast materials database to identify candidates for spintronic applications in computing and smartphones.
A new framework for portfolio management uses deep reinforcement learning to predict price trends and make strategic decisions, overcoming limitations of existing systems. The system consists of evolving agent modules and strategic agent modules, allowing for modular design and scalability.
A study analyzed Reddit posts to identify common parenting concerns, including sleep, eating, and emotional difficulties. The authors used latent Dirichlet allocation to categorize the posts and found that these topics are consistently discussed among parents.
A Penn State study found that live-streaming ads for alcohol, energy drinks, and junk food significantly increased during the pandemic, targeting young viewers. The researchers discovered that energy drinks made up nearly 80% of marketing, while ads for restaurants and soda were also prevalent.
Rice University computer scientists have discovered an inexpensive way to implement rigorous personal data privacy in large databases for machine learning. Using locality sensitive hashing, their RACE method creates small summaries of enormous databases while scaling for high-dimensional data.
Researchers at UVA Health System found that fluoxetine, an antidepressant commonly known as Prozac, shows promise against age-related macular degeneration. Patients taking fluoxetine were less likely to develop dry AMD, a condition affecting nearly 200 million people worldwide.
Researchers at GlaxoSmithKline and CCDC combined proprietary and published datasets to train machine learning models for predicting stable polymorphs in new drug candidates. The approach leverages the large volume and variety of data in the Cambridge Structural Database, resulting in more confident predictions and improved model accuracy.
A Mount Sinai study found that polygenic risk scores were no better at predicting worsening symptoms than written reports in schizophrenia patients. The results raise questions about the use of polygenic risk scores in real-world situations, suggesting a doctor's report may be an untapped source of predictive information.
A new algorithm, Phe2vec, accurately identified patients with certain diseases, outperforming traditional methods in classifying diagnoses. The study suggests that this automation will facilitate further research in clinical informatics.
The new search engine enables researchers to compare different samples with the same function in various microbiomes, aiding disease diagnosis and treatment. MSE 2 includes an extended database of over 260,000 samples, providing a bird's eye view of the microbiome data space
The study uses Synthetic Aperture Radar Interferometry (InSAR) to investigate lignite regions in North Rhine-Westphalia, Germany. It finds significant land subsidence in open-cast mining areas, with rates of up to 50 centimeters per year.
A new study warns that massive African reforestation efforts could harm ecosystems, as the biomes are divided into distinct types with unique tree species. The researchers analyzed 753 sites in both environments, finding that rainfall, seasonality, and temperature are key environmental factors.
The study aims to examine the extent to which the AD&A Second Major programme equips students with key skills and competencies in the future accounting workplace. It will also measure students' learning outcomes and gather insights on how future programmes can be designed to equip students with necessary skills.
KDD 2020 features four keynote talks on meta-provenance, AI for intelligent financial services, state-space multi-taper time-frequency analysis, and computational epidemiology. The conference will also include 18 applied data science invited talks and 217 accepted research papers.
Researchers found nearly three-fourths of foreshocks preceded mainshocks by days to weeks, significantly higher than previously understood. Advanced signal processing techniques and computing capabilities enabled the detection of small foreshocks with magnitudes less than 1.
The KDD Cup 2019 competition featured three tracks tackling societal challenges such as transportation and malaria. Teams won prizes of up to $15,000 by applying machine learning tools to complex problems.
Researchers developed a computational approach to design new hybrid halide semiconductors, discovering 13 new material candidates for solar cells and 23 for LEDs. The study used large-scale quantum mechanics calculations to identify novel materials with excellent optoelectronic properties.
A two-day conference brought together experts in AI, policy, law enforcement, and survivors to tackle human trafficking. Researchers aim to leverage technology to support trained law enforcement in tackling this complex issue, combining data analysis with social policy and political efforts.
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 team at University of Washington demonstrates an innovative approach to bridge AFM and big data, offering better spatial resolution and accuracy. By using sequential excitation strategy, they deduce physical insight from PCA data and speed up analysis by orders of magnitude.
The Microbiome Search Engine (MSE) enables fast comparison of microbiomes by searching for structurally or functionally similar microbes. MSE proposes innovative evaluation indices to measure microbiome novelty and impact, providing a bird's eye view of the microbiome data universe.
A new text-mining method developed by Cornell statistician Shawn Mankad and his colleagues can help developers improve mobile apps faster. By aggregating and parsing customer reviews in one step, the method provides guidance on a single app's performance and compares it to competing apps over time.
A team of researchers has developed a mutual information approach to interpreting atmospheric data collected over an 18-year period, finding strong correlations between new-particle formation and water content, sulfuric acid concentration, temperature, and relative humidity.
This book by East China Normal University and Singapore Management University researchers delves into network data mining and analysis to identify social communities, assess network robustness, and predict links. The research focuses on bipartite graphs and signed graphs, offering insights into user behavior across heterogeneous networks.
A new web-based mapping tool shows the land laid bare by mountaintop coal mining in central Appalachia each year, going back more than three decades. The tool estimates that between 1985 and 2015, an average of 21,000 acres was converted to bare earth and rubble annually.