Andrew Schroeder’s Post

What does flourishing look like for scientists, science communities, and society? Join the discussion by submitting an abstract to our interdisciplinary session, "Flourishing Science Commons: Data Science, Open Science, and Knowledge Communities" at the American Geophysical Union meeting this December! The session is co-convened by Ryan McGranaghan, Chelle Gentemann, Lisa Cuevas Shaw, PhD and myself. Check out the abstract solicitation below and consider joining us! We are seeking many perspectives on the future of scientific communities and collaboration. #AGU2024 #science #metascience

⭐ 𝐄𝐱𝐜𝐢𝐭𝐞𝐝 𝐭𝐨 𝐒𝐡𝐚𝐫𝐞 𝐚 𝐒𝐢𝐠𝐧𝐢𝐟𝐢𝐜𝐚𝐧𝐭 𝐀𝐜𝐡𝐢𝐞𝐯𝐞𝐦𝐞𝐧𝐭! ⭐ I'm happy to share that my latest Research paper, titled "𝐋𝐚𝐧𝐝𝐬𝐥𝐢𝐩 𝐓𝐫𝐢𝐠𝐠𝐞𝐫 𝐂𝐥𝐚𝐬𝐬𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐆𝐥𝐨𝐛𝐚𝐥 𝐋𝐚𝐧𝐝𝐬𝐥𝐢𝐝𝐞 𝐂𝐚𝐭𝐚𝐥𝐨𝐠 𝐃𝐚𝐭𝐚𝐬𝐞𝐭 𝐔𝐬𝐢𝐧𝐠 𝐌𝐚𝐜𝐡𝐢𝐧𝐞 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐀𝐥𝐠𝐨𝐫𝐢𝐭𝐡𝐦," has been published in IEEE Xplore! In this study, we delve into the innovative use of machine learning algorithms to enhance the classification of landslip triggers. This breakthrough aims to significantly improve public safety and operational efficiency in various environments. I sincerely thank my co-author, Dr. F. Sangeetha Francelin Vinnarasi, Professor, Department of Information Technology, St. Joseph's Institute of Technology, who played a major role in this work and supported this journey. 📄 Access the full paper here https://lnkd.in/gFxduyAB #Research #MachineLearning #Landslide #PublicSafety #IEEE #Innovation #Collaboration

🚀 Exciting Research Alert! 📄 We are thrilled to announce the publication of our latest research paper titled "D-CODE: Data Colony Optimization for Dynamic Network Efficiency" in the field of Neural and Evolutionary Computing. 🔍 Our paper introduces D-CODE, a novel framework that blends Data Colony Optimization (DCO) algorithms with Dynamic Efficiency (DE) models. This innovative approach is inspired by the collective behaviors of biological colonies and is designed for real-time adaptation in complex data landscapes. 💡 Key Highlights: - Robust Optimization: Utilizes metaheuristic strategies from ant colonies, bee swarms, and fungal networks. - Dynamic Responsiveness: Enables continuous resource recalibration and process adjustments amidst changing conditions. - Empirical Validation: Demonstrates improvements in solution quality, faster convergence rates, and higher computational efficiency. Huge thanks to Ayush Thakur, my co-author, for his invaluable contributions to our research paper. His expertise and dedication elevated the quality of our work. 🌐 Read our paper for a deep dive into how D-CODE can transform intelligent systems design and operational efficiency. We're excited about the potential applications and the promising outcomes of our research. 🔗 Find the paper here: https://lnkd.in/g3bAyNSb #Research #Innovation #DataScience #NeuralComputing #EvolutionaryComputing #DCODE #Optimization #DynamicEfficiency #ComputationalIntelligence #NeuralNetwork

Preprint and post-publication review, full data and methodology disclosure, rigorous COI policies, and uncoupling editorship and publication decisions are all absolutely necessary for the sustainability of peer review in science. The old model is broken.

🚀 Hot from the press: GHOST - Graph-based Higher-Order Similarity Transformation for Classification (Pattern Recognition (IF:8), Elsevier) 🔗Publication available (open access) here : https://lnkd.in/e2JeStax 🧠 Well done 💪Enzo Battistella (Broad Institute of MIT and Harvard)!!! 🎯with contributions from Maria Vakalopoulou (CentraleSupélec, Université Paris-Saclay, Inria), Nikos Paragios (TheraPanacea, CentraleSupélec, Université Paris-Saclay) & eric deutsch (Gustave Roussy, INSERM, Université Paris-Saclay) 🔍 This paper explores new frontiers in feature representation for high-dimensional datasets. Despite the abundance of feature selection techniques and distance metrics, pinpointing the optimal choice for a given task remains complex. 📄 GHOST is an innovative method designs high-order distance metrics utilizing a sparse selection of features specifically for classification. By leveraging Conditional Random Field (CRF) energy minimization and Dual Decomposition, our approach ensures both efficiency and flexibility in the features considered. 🔬 Rigorous evaluation: Our method has been meticulously tested on synthetic data and for Covid-19 patient stratification. Through comparative analysis with state-of-the-art baselines, our proposed method demonstrates superior performance and relevance of the learned metric. 💡 This work redefines feature representation and distance metric design, pushing the boundaries of what's possible in the field of data science. #DataScience #MachineLearning #FeatureSelection #HighDimensionalData #CRF #graphicalmodels #metriclearning #manifoldlearning

My research paper was recently published in a prestigious, international scientific journal, IEEE Xplore. The title of my work is: Accuracy and Performance of Machine Learning Methodologies: Novel Assessments of Country Pandemic Vulnerability Based on Non-Pandemic Predictors. I am also very proud to announce Professor Dr. Simske as a co-author. This paper analyzes two novel public health indices, Population Health Index (PHI) and Country Health Index (CHI), to provide a novel scoring model for country level pandemic risk assessment. https://lnkd.in/ec3wQUbc

📃Scientific paper: The social graph based on real data Abstract: In this paper, we propose a model enabling the creation of a social graph corresponding to real society. The procedure uses data describing the real social relations in the community, like marital status or number of kids. Results show the power-law behavior of the distribution of links and, typical for small worlds, the independence of the clustering coefficient on the size of the graph. ;Comment: Preprint of the paper presented at International Conference on Computational Science, ICCS 2023, Prague Continued on ES/IODE ➡️ https://etcse.fr/5bQ ------- If you find this interesting, feel free to follow, comment and share. We need your help to enhance our visibility, so that our platform continues to serve you.

– Physical Immortality ? There´re people linked to Scientific Community who believe in it in the future. This is a very important subject in this post, of course. https://lnkd.in/dBYY2iUV

🚀 Building the Knowledge Base of Tomorrow: Our Systems Started Crawling the Scientific Landscape! One of the most significant challenges of combining scientific research and artificial intelligence is the amount of data to be handled. To give you a sense of scale, around 3 million scientific articles are published each year—that is over 8,000 articles per day! At first sight, that might not seem like a lot. Analyzing 8,000 articles daily is undoubtedly an impossible task for a single researcher. For us, it totals to petabytes of raw data that need to be analyzed but, most importantly, discovered at first. After extensive research—and loads of optimization—we have put our experimental "discovery cluster" into operation. After this test run, we plan to launch the final version of our cluster, which is expected to index the majority of scientific literature in a matter of weeks 🎉 . 💭 Current estimates say there are around 250 million scientific articles in existence. We ourselves are very interested in how much larger (or smaller) the final number will turn out. Do you agree with the estimate or have a different number in mind? Feel free to comment. 🔔 Follow us for more updates

Dr. Yimeng Song, postdoctoral research associate at Yale School of the Environment, is joining DDL next week for a presentation on "Leveraging Geospatial Data Fusion and Computational Methods for Environmental Exposure and Health Assessment." Abstract: Climate change and urbanization are major environmental challenges that have significant impacts on human health. Understanding how people are exposed to these changing environments is crucial for reducing health inequalities. In this context, it is essential to provide scientific evidence that offers a comprehensive understanding of spatial and temporal dynamics of human-environment interactions. Advances in multi-source geospatial big data, combined with innovative computational methods such as geospatial big data fusion, automated machine learning, and cloud computing enable detailed investigation into the dynamic interactions between humans and environments. This talk will introduce our recent studies on leveraging multi-source geospatial big data and computational methods to measure and analyze human mobility, environmental exposure and injustice, and environmental their health effects, aiming to uncover the complex relationships between environments and health. Our findings highlight the potential of these innovative computational methods to inform environmental management strategies and public health policies, ultimately contributing to more sustainable and healthy communities. 📅 July 31 🕐 1:15 PM ET 🔗https://lnkd.in/eBRkyKxd