Article
Biochemical Research Methods
Alfred Ultsch, Joern Loetsch
Summary: In bioinformatics data processing, data transformations are commonly used for data projection and clustering. However, the commonly used Euclidean distance metric is not scale invariant and may be inappropriate for complex variables, leading to negative impacts on cluster analysis results. This study proposes the EDO transformation as a better alternative to traditional z-standardization, and demonstrates its effectiveness through simulations and real data applications.
BMC BIOINFORMATICS
(2022)
Article
Multidisciplinary Sciences
Afifa Salsabil Fathima, Syed Muzamil Basha, Syed Thouheed Ahmed, Sandeep Kumar Mathivanan, Sukumar Rajendran, Saurav Mallik, Zhongming Zhao
Summary: This article presents a conceptual framework for the analysis and uniformization of medical datasets using Federated Learning. By utilizing multi-tier neural networks for feature synchronization and parameter extraction, the study achieved dataset standardization and labeling with high accuracy.
Article
Biochemical Research Methods
Umberto Ferraro Petrillo, Francesco Palini, Giuseppe Cattaneo, Raffaele Giancarlo
Summary: The study introduces a new big data platform, FADE, for alignment-free genomic analysis, supporting 18 best-performing AF functions, with faster execution time and user-friendly software design. Additionally, it provides a novel analysis of the informativeness and robustness of AF functions, finding that only a handful of functions out of the 18 included in FADE can actually be used.
Review
Engineering, Biomedical
Jacob Kerner, Alan Dogan, Horst von Recum
Summary: Machine learning has been widely utilized in various fields, including biomaterials, optimizing data collection and analysis. Recent advances in biomaterials have focused on quantitative structure properties relationships, introducing four basic models for rapid development and addressing the lack of machine learning implementation in the field. This article aims to spark greater interest and awareness in utilizing computational methods for biomaterials research.
ACTA BIOMATERIALIA
(2021)
Article
Engineering, Civil
Qiao Dong, Xueqin Chen, Shi Dong, Fujian Ni
Summary: This paper summarized over 40 data analysis methods used in pavement engineering, including statistical tests, regression, and machine learning. Traditional statistical regression models are suitable for quantifying significant factors and predicting pavement performance, while supervised machine learning is effective for dealing with large or unstructured data.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Computer Science, Information Systems
En-Hau Yeh, Phone Lin, Ming-Wey Huang
Summary: This article proposes an anomaly detection framework based on population distribution, using mobile network log data to monitor real-time population mobility patterns and identify critical indicators for sudden events. The framework shows a high practicality in actual situations, as demonstrated by the experiments conducted during the 2018 Hualien Earthquake in Taiwan.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Medicine, General & Internal
Jenna M. Reps, Patrick Ryan, P. R. Rijnbeek
Summary: The study aimed to quantifying the generalisability of prediction models by investigating the impact of different development and internal validation designs in big data. Results showed that even with large data, models tend to overfit without a proper validation process. Validation processes to select hyperparameters and assess internal validation are crucial for accurate model performance evaluation.
Article
Hospitality, Leisure, Sport & Tourism
Adam Weaver
Summary: Efforts to aggregate data comprehensively have led to a crisis of analysis in the tourism industry, where individuals are increasingly treated as mere objects. The recognition of tourism as a series of distinctive human actions is overshadowed by a fervor for impersonal mass quantification, reflecting tensions caused by a positivistic, business-driven way of knowing in the industry.
ANNALS OF TOURISM RESEARCH
(2021)
Article
Geochemistry & Geophysics
Fantine Huot, R. Lily Hu, Nita Goyal, Tharun Sankar, Matthias Ihme, Yi-Fan Chen
Summary: This article introduces a large-scale historical wildfire dataset that combines various explanatory variables and uses machine learning methods to predict wildfire spread. This dataset can serve as a benchmark for developing proactive wildfire propagation models based on remote-sensing data within a one-day lead time.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Review
Biochemistry & Molecular Biology
Madeline Alizadeh, Natalia Sampaio Moura, Alyssa Schledwitz, Seema A. Patil, Jacques Ravel, Jean-Pierre Raufman
Summary: Studying individual data types in isolation is limited in providing comprehensive answers, but multi-omics approaches can generate and integrate multiple data types to offer a holistic understanding of biological and disease processes. Gastroenterology and hepatobiliary research benefit from these approaches due to the interconnectedness of the GI tract, brain, immune and endocrine systems, and GI microbiome. The use of big data in multi-omic, multi-site studies allows for better investigations into the connections between organ systems and more accurate evaluations of interventions.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Article
Chemistry, Multidisciplinary
Lauren Takahashi, Thanh Nhat Nguyen, Sunao Nakanowatari, Aya Fujiwara, Toshiaki Taniike, Keisuke Takahashi
Summary: Research on designing high-performance catalysts for the oxidative coupling of methane is often hindered by inconsistent data, but high throughput experiments provide a systematic way to produce catalyst-related data. By applying graph theory to visualize trends in data transformation, new catalysts can be designed to achieve high C-2 yields, resulting in the successful design of numerous efficient catalysts.
Article
Electrochemistry
Karthik S. Mayilvahanan, Kenneth J. Takeuchi, Esther S. Takeuchi, Amy C. Marschilok, Alan C. West
Summary: Models that understand and predict degradation play a crucial role in improving the performance of Li-ion batteries. This study combines physical mechanistic models with machine learning to analyze synthetic low rate charge curves and study different thermodynamic degradation modes. Establishing interpretable machine learning models through step-by-step procedures, including data set splitting, featurization, and model fitting for regression and classification tasks.
BATTERIES & SUPERCAPS
(2022)
Review
Biochemical Research Methods
Mohamed Nadif, Francois Role
Summary: Biomedical scientific literature is growing rapidly, making it challenging to identify relevant results; automated information extraction tools based on text mining techniques are essential; deep neural networks have significantly advanced this research field.
BRIEFINGS IN BIOINFORMATICS
(2021)
Review
Biochemistry & Molecular Biology
Romano Weiss, Sanaz Karimijafarbigloo, Dirk Roggenbuck, Stefan Roediger
Summary: Neural networks, also known as artificial neural networks, are crucial tools in deep-learning applications, and their popularity has soared since the early 2000s. This review focuses on the use of deep learning in biomedical data analysis, particularly in the analysis of biomarkers in bioimage data. The article also provides quantitative insights into the usage of network types in different scientific fields based on a data analysis of neural network publications.
Article
Automation & Control Systems
Zeyu Yang, Zhiqiang Ge
Summary: The arrival of the intelligent manufacturing and industrial internet era has brought opportunities and challenges to modern industry. Industrial Big Data analytics, as the core link between intelligent manufacturing and industrial internet platform, has received increasing attention. The efficient mining of high-value information and the utilization of real-life industrial process are hot topics. With the development of industrial automation, the learning paradigm of industrial Big Data analytics is evolving accordingly.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)