Generated Data Sets Research Articles

Artificial intelligence based data curation: enabling a patient-centric European health data space.

The emerging European Health Data Space (EHDS) Regulation opens new prospects for large-scale sharing and re-use of health data. Yet, the proposed regulation suffers from two important limitations: it is designed to benefit the whole population with limited consideration for individuals, and the generation of secondary datasets from heterogeneous, unlinked patient data will remain burdensome. AIDAVA, a Horizon Europe project that started in September 2022, proposes to address both shortcomings by providing patients with an AI-based virtual assistant that maximises automation in the integration and transformation of their health data into an interoperable, longitudinal health record. This personal record can then be used to inform patient-related decisions at the point of care, whether this is the usual point of care or a possible cross-border point of care. The personal record can also be used to generate population datasets for research and policymaking. The proposed solution will enable a much-needed paradigm shift in health data management, implementing a 'curate once at patient level, use many times' approach, primarily for the benefit of patients and their care providers, but also for more efficient generation of high-quality secondary datasets. After 15 months, the project shows promising preliminary results in achieving automation in the integration and transformation of heterogeneous data of each individual patient, once the content of the data sources managed by the data holders has been formally described. Additionally, the conceptualization phase of the project identified a set of recommendations for the development of a patient-centric EHDS, significantly facilitating the generation of data for secondary use.

A Survey on Automatic Generation of Figurative Language: From Rule-based Systems to Large Language Models

Figurative language generation (FLG) is the task of reformulating a given text to include a desired figure of speech, such as a hyperbole, a simile, and several others, while still being faithful to the original context. This is a fundamental, yet challenging task in Natural Language Processing (NLP), which has recently received increased attention due to the promising performance brought by pre-trained language models. Our survey provides a systematic overview of the development of FLG, mostly in English, starting with the description of some common figures of speech, their corresponding generation tasks, and datasets. We then focus on various modelling approaches and assessment strategies, leading us to discussing some challenges in this field, and suggesting some potential directions for future research. To the best of our knowledge, this is the first survey that summarizes the progress of FLG including the most recent development in NLP. We also organize corresponding resources, e.g., article lists and datasets, and make them accessible in an open repository. We hope this survey can help researchers in NLP and related fields to easily track the academic frontier, providing them with a landscape and a roadmap of this area.

Prediction and Analysis of Transient Turbine Tip Clearance Using Long Short-Term Memory Neural Network

Abstract The transient turbine tip clearance (δ) throughout the engine process is crucial to modern high-performance aero engines. However, there is still a lack of efficient and accurate transient prediction models of tip clearances with active thermal control (ATC) system, especially for the tip clearances of the complex turbine structures with various parameters. This study develops a transient prediction model for the tradeoff between computational efficiency and accuracy, which includes an offline dataset generation process and an online δ prediction process. The offline dataset is first generated using an in-house finite element analysis code, which is validated against a transient tip clearance experiment, and data splicing and sensitivity analysis are applied to enrich the sample features and reduce the input parameters' dimensionality. Then, the long short-term memory neural network (LSTM) is employed to learn the transient tip clearances' timing information. The time consumption for the transient prediction model is significantly shorter than that for the tip clearance calculation method by three orders, and the maximum relative error is as low as 3.59%. In addition, the transient characteristics, including the overshoot value (σ) and the response time (ts), are investigated with different jet Reynolds numbers (Rec) and temperatures (Tfc) of ATC cooling flow. The ts decreases with larger Rec and smaller Tfc due to a more significant cooling effect. However, the σ increases with the increase of Rec and Tfc due to the different sensitivity of cooling parameters. This study provides a reference for the transient tip clearance prediction and the adjustments in the cooling strategies.

Analyzing the Merits and Drawbacks of Multilingual Text Detection Datasets for Videos

In recent years, video contents have observed a massive growth with online lecture, news channels, and media contents and with short clips. These contents have objects, texts in them. The majority of the time, multilingual datasets are not available as per the need and have to utilize YouTube videos with multiple languages as a source . This study proposes a systematic structure for the generation of multilingual datasets and explains the purpose for their creation. Additionally, the study demonstrated the difficulties in creating the data set, like segregating data that is multilingual, etc. Lastly, discusses the effects of creating multilingual datasets because of various difficulties. As a result, the research's contribution is the multilingual dataset that was created using the suggested systematic structure, which also practically shows its limitations and benefits. Key Words:, Data Set, Multilingual, Text detection, challenge, consequences

Unsupervised recognition of radar signals combining multi-block TFR with subspace clustering

In the realm of radar systems, the proliferation of new modulation techniques introduces an increased level of complexity in the identification of both existing and potentially novel modulation schemes. Conventional supervised recognition methodologies, which rely heavily on labeled datasets, exhibit limitations in distinguishing amongst a multitude of unlabeled signals. This paper introduces a novel framework that synergizes Subspace Clustering (SC) with Multi-Block Time-Frequency Representations (TFRs, denoted as mT), referred to as SC-mT. This framework leverages subspace clustering for the unsupervised categorization of radar modulations and employs an innovative multi-block strategy to augment classification precision. The process commences with the generation of TFR datasets for radar signals, followed by the construction of two distinct multi-block models, Model-A and Model-B. These models segment the radar TFRs into overlapping multi-block sets, utilizing the concept of random receptive fields. The subspace clustering algorithm is then applied to each block within the sub-TFR sets to procure an affinity matrix. The culmination of this process involves the aggregation of the affinity matrices from all blocks, facilitating the derivation of classification outcomes via spectral clustering. Empirical analyses affirm that the sequential integration of the SC-mT algorithm surpasses the classification efficacy of various conventional and cutting-edge algorithms. Notably, this algorithm attains a classification accuracy surpassing 90% for ten unlabeled signals, even in scenarios where the signal-to-noise ratio (SNR) is as low as -4 dB.

Broadening the horizon of immune thrombocytopenia through Omics approaches.

Immune thrombocytopenia (ITP) is a highly heterogeneous autoimmune bleeding disorder characterized by low platelet counts due to an immune-mediated platelet destruction and impaired platelet production. The pathophysiology is multifactorial and remains to be fully unravelled. Consequently, disease trajectories and responses to therapeutics, despite the availability of multiple agents, can be unpredictable and differing between patients. There is an urgent need for the identification of diagnostic and therapeutic biomarkers, but this has proven to be challenging to achieve. To shed light on this, two studies in this issue of the British Journal of Haematology have recognized the opportunity of using high-throughput Omics technologies in ITP. Sun etal. performed proteomics, and Li etal. metabolomics, on bone marrow biopsy samples of patients with ITP. This was conducted using mass spectrometry and, due to the generation of large datasets, in combination with machine learning. These studies set the stage for further investigations exploring the high potential of multi-omics technologies in order to shed light on the heterogeneity in ITP, accelerating the path towards a much needed personalized medicine approach. Commentary on: Li etal. Metabolomics profile and machine learning prediction of treatment responses in immune thrombocytopenia: A prospective cohort study. Br J Haematol 2024;204:2405-2417. Commentary on: Sun etal. Proteomics landscape and machine learning prediction of long-term response to splenectomy in primary immune thrombocytopenia. Br J Haematol 2024;204:2418-2428.

Material Flow Analysis and Occupational Exposure Assessment in Additive Manufacturing End-of-Life Material Management.

Additive manufacturing (AM) offers a variety of material manufacturing techniques for a wide range of applications across many industries. Most efforts at process optimization and exposure assessment for AM are centered around the manufacturing process. However, identifying the material allocation and potentially harmful exposures in end-of-life (EoL) management is equally crucial to mitigating environmental releases and occupational health impacts within the AM supply chain. This research tracks the allocation and potential releases of AM EoL materials within the US through a material flow analysis. Of the generated AM EoL materials, 58% are incinerated, 33% are landfilled, and 9% are recycled by weight. The generated data set was then used to examine the theoretical occupational hazards during AM EoL material management practices through generic exposure scenario assessment, highlighting the importance of ventilation and personal protective equipment at all stages of AM material management. This research identifies pollution sources, offering policymakers and stakeholders insights to shape pollution prevention and worker safety strategies within the US AM EoL management pathways.

Automating Wood Species Detection and Classification in Microscopic Images of Fibrous Materials with Deep Learning.

We have developed a methodology for the systematic generation of a large image dataset of macerated wood references, which we used to generate image data for nine hardwood genera. This is the basis for a substantial approach to automate, for the first time, the identification of hardwood species in microscopic images of fibrous materials by deep learning. Our methodology includes a flexible pipeline for easy annotation of vessel elements. We compare the performance of different neural network architectures and hyperparameters. Our proposed method performs similarly well to human experts. In the future, this will improve controls on global wood fiber product flows to protect forests.

Inference of dynamic hypergraph representations in temporal interaction data.

A range of systems across the social and natural sciences generate data sets consisting of interactions between two distinct categories of items at various instances in time. Online shopping, for example, generates purchasing events of the form (user, product, time of purchase), and mutualistic interactions in plant-pollinator systems generate pollination events of the form (insect, plant, time of pollination). These data sets can be meaningfully modeled as temporal hypergraph snapshots in which multiple items within one category (i.e., online shoppers) share a hyperedge if they interacted with a common item in the other category (i.e., purchased the same product) within a given time window, allowing for the application of hypergraph analysis techniques. However, it is often unclear how to choose the number and duration of these temporal snapshots, which have a strong influence on the final hypergraph representations. Here we propose a principled nonparametric solution to this problem by extracting temporal hypergraph snapshots that optimally capture structural regularities in temporal event data according to the minimum description length principle. We demonstrate our methods on real and synthetic data sets, finding that they can recover planted artificial hypergraph structure in the presence of considerable noise and reveal meaningful activity fluctuations in human mobility data.

Automated Text Recognition and Segmentation for Historic Map Vectorization: A Mask R-CNN and UNet Approach

Historic maps are essential for comprehending how buildings and landscapes have changed over time. For this—vectorization can be a useful method of analysis for an extensive collection of these maps. However, text overlaps with structural elements—often makes this process more difficult. Therefore, an automated pipeline for text recognition, pixel-level text mask creation, dataset generation, and text bounding box detection has been proposed. Findings shows—text segmentation, detection, and recognition were demonstrated by the combination of Mask Region-based Convolutional Neural Network (Mask R-CNN) and UNet model achieved a 99.12% of all text occurrences in images—which also attained an accuracy of 87.72% while collecting text inside bounding boxes. This end-to-end pipeline shows potential for a wide range of future uses, especially when it comes to text removal for the purpose of making historic maps easier to vectorize and analyze—which will improve the understanding of historical buildings and landscapes.

Drought Monitoring Using Moderate Resolution Imaging Spectroradiometer-Derived NDVI Anomalies in Northern Algeria from 2011 to 2022

Drought has emerged as a major challenge to global food and water security, and is particularly pronounced for Algeria, which frequently grapples with water shortages. This paper sought to monitor and assess the temporal and spatial distribution of drought severity across northern Algeria (excluding the Sahara) during the growing season from 2011 to 2022, while exploring the relationship between the normalized difference vegetation index (NDVI) anomaly and climate variables (rainfall and temperature). Temporal NDVI data from the Terra moderate resolution imaging spectroradiometer (MODIS) satellite covering the period 2000–2022 and climate data from the European Reanalysis 5th Generation (ERA5) datasets collected during the period 1990–2022 were used. The results showed that a considerable portion of northern Algeria has suffered from droughts of varying degrees of severity during the study period. The years 2022, 2021, 2016, and 2018 were the hardest hit, with 76%, 71%, 66%, and 60% of the area, respectively, experiencing drought conditions. While the relationship between the NDVI anomaly and the climatic factors showed variability across the different years, the steady decrease in vegetation health indicated by the NDVI anomaly corroborates the observed increase in drought intensity during the study period. We conclude that the MODIS-NDVI product offers a cost-efficient approach to monitor drought in data-scarce regions like Algeria, presenting a viable alternative to conventional climate-based drought indices, while serving as an initial step towards formulating drought mitigation plans.

CLIPPER 2.0: Peptide-Level Annotation and Data Analysis for Positional Proteomics

Positional proteomics methodologies have transformed protease research, and have brought mass spectrometry (MS)-based degradomics studies to the forefront of protease characterization and system-wide interrogation of protease signaling. Considerable advancements in both sensitivity and throughput of liquid chromatography (LC)-MS/MS instrumentation enable the generation of enormous positional proteomics datasets of natural and protein termini and neo-termini of cleaved protease substrates. However, concomitant progress has not been observed to the same extent in data analysis and post-processing steps, arguably constituting the largest bottleneck in positional proteomics workflows. Here, we present a computational tool, CLIPPER 2.0, that builds on prior algorithms developed for MS-based protein termini analysis, facilitating peptide-level annotation and data analysis. CLIPPER 2.0 can be used with several sample preparation workflows and proteomics search algorithms and enables fast and automated database information retrieval, statistical and network analysis, as well as visualization of terminomic datasets. We demonstrate the applicability of our tool by analyzing GluC and MMP9 cleavages in HeLa lysates. CLIPPER 2.0 is available at https://github.com/UadKLab/CLIPPER-2.0.

Unified predictive modeling of supercontinuum spectra: Using multi-material data with Closed-Form Continuous Time Neural Networks

This paper explores the application of the Closed-Form Continuous-Time Neural Networks (CfC) model in predicting supercontinuum spectra in a unified dataset of various core materials in a planar waveguide. Through numerical simulations and dataset generation, the study constructs a robust model capable of accurately predicting spectral behavior under different conditions and material variation. Initially, the unified dataset comprises spectral data for three materials: Silicon Nitride (Si3N4), Lithium Niobate (LiNbO3), and Silicon Carbide (SiC). The CfC model demonstrates remarkable accuracy in capturing spectral nuances and exhibits a minimal Mean Squared Error (MSE) loss value of 7.2537×10−4. Subsequently, the dataset is expanded to include Tantalum Pentoxide (Ta2O5), introducing a fourth material for evaluation. The inclusion of Ta2O5 data further validates the model’s scalability and generalization capabilities with Mean Squared Error (MSE) loss value of 6.7894×10−4. The advantage of the CfC model in precisely predicting supercontinuum spectra while retaining computational efficiency is demonstrated by comparisons with conventional methods. The CfC model is a viable tool for enhancing predictive modeling in photonics research because of its scalability and generalization characteristics. This will pave the way for future developments in deep learning techniques for optical device optimization.

Modelling the vulnerability of urban settings to wildland–urban interface fires in Chile

Abstract. Wildland–urban interface (WUI) regions are particularly vulnerable to wildfires due to their proximity to both nature and urban developments, posing significant risks to lives and property. To enhance our understanding of the risk profiles in WUI areas, we analysed seven fire case studies in central Chile. We developed a mixed-method approach for conducting local-scale analyses, which involved field surveys, remote-sensing through satellite and drone imagery, and GIS-based analysis of the collected data. The methodology led to the generation of a georeferenced dataset of damaged and undamaged dwellings, including 16 variables representing their physical characteristics, spatial arrangement, and the availability of fire suppression resources. A binary classification model was then used to assess the relative importance of these attributes as indicators of vulnerability. The analysis revealed that spatial arrangement factors have a greater impact on damage prediction than the structural conditions and fire preparedness of individual units. Specifically, factors such as dwelling proximity to neighbours, distance to vegetation, proximity to the border of dwelling groups, and distance from the origin of the fire substantially contribute to the prediction of fire damage. Other structural attributes associated with less affluent homes may also increase the likelihood of damage, although further data are required for confirmation. This study provides insights for the design, planning, and governance of WUI areas in Chile, aiding the development of risk mitigation strategies for both built structures and the broader territorial area.

Empowering Global AMR Research Community: Interactive GIS dashboards for AMR data analysis and informed decision-making

Background Antimicrobial Resistance (AMR) is a critical global public health concern, demanding effective tools for research, data analysis, and decision-making. This study proposes a groundbreaking approach to empower the global AMR research community by introducing interactive Geographic Information System (GIS) dashboards. These dashboards aim to facilitate comprehensive data analysis of AMR across multiple countries, providing insights into antimicrobial usage (AMU), resistance patterns, and geographic distribution. Methods The novel approach involves the development of GIS dashboards that integrate and harmonize data from diverse sources, including clinical laboratories, surveillance networks, and public health agencies. Objective 1 focuses on creating a dashboard encompassing all countries, offering comprehensive data analysis capabilities and visualization tools. Objective 2 entails building a focused dashboard specifically for Kenya and Uganda, allowing for comparative analysis of AMR in these regions. Objective 3 involves the generation of a simulated dataset for Kampala, Uganda, addressing data limitations in that specific area. Results The GIS dashboards serve as powerful tools for visualizing and analyzing AMR-related datasets, providing stakeholders with a comprehensive view of the global AMR landscape. These dashboards offer valuable insights into antimicrobial usage, resistance patterns, and geographical distribution. The centralized platform facilitates data exploration and analysis, aiding researchers, policymakers, and healthcare professionals in making informed decisions to combat AMR. Conclusions In conclusion, the study demonstrates that the developed GIS dashboards empower stakeholders by providing valuable insights and informed decision-making capabilities. The dashboards serve as essential tools for addressing the global challenge of AMR, allowing for a deeper understanding of the problem and informing effective strategies. The approach outlined in this study has the potential to significantly contribute to the ongoing efforts to combat AMR on a global scale.

A time-saving fault diagnosis using simplified fast GAN and triple-type data transfer learning

Existing intelligent fault diagnosis approaches demand substantial data for training diagnostic models. However, factors such as the inherent characteristics of bearings, operating conditions, and privacy security make collecting comprehensive fault-bearing data very difficult. Although generating synthetic data through generative adversarial networks (GANs) is feasible, the data generation of GANs is a time-consuming process. To address these challenges, a fault diagnosis framework based on GAN and deep transfer learning (DTL) is proposed, termed the simplified fast GAN triple-type data transfer learning (SFGAN-TDTL) method. Initially, an SFGAN is proposed as a replacement for traditional GANs. The time-frequency image data generated by SFGAN serve to augment the training dataset, offering faster and higher-quality data generation compared to traditional GANs. To further reduce time consumption for GAN-based methods, the TDTL method is proposed. Differing from DTL, which utilizes synthetic data to construct a pre-trained model and conducts targeted fine-tuning with real data, TDTL employs open-source data, synthetic data, and real data to fill the weights of the task-insensitive layer, task-sensitive layer, and fully connected layer, respectively. Numerical results demonstrate that SFGAN-TDTL maintains higher diagnostic accuracy while significantly reducing time consumption.

Metaheuristic algorithms for solving roman {2}-domination problem

A Roman {2}-dominating function (Rom2DF) on a graph G(V, E) is a function g : V → {0, 1, 2} of G such that for every vertex x ∈ V with g(x) = 0, either there exists a neighbor y of x with g(y) = 2 or at least two neighbors, u, v with g(u) = g(v) = 1. The value w(g) = ∑x∈V g(x) is the weight of the Rom2DF. The minimum weight of a Rom2DF of G is called the Roman {2}-domination number denoted by γ{R2}(G). Since determining γ{R2}(G) of a graph G is NP-hard and no metaheuristic algorithms have been proposed for the same, two procedures based on genetic algorithm are proposed as a solution for the Roman {2}-domination problem. One of the proposed methods employs a random initial population, while the other uses a population generated using heuristics. Experiments have been carried out on graphs generated using Erdös–Rényi model, a popular model for graph generation and Harwell Boeing (HB) dataset. The experimental results demonstrate that both approaches provide a near optimal solution which is well within the known lower and upper bounds for the problem. The experimental results further show that the procedure based on random initial population has outperformed the heuristic based procedure.

Modelling of aero-mechanical response of wind turbine blade with damages by computational fluid dynamics, finite element analysis and Bayesian network

Computational Fluid Dynamics and Finite Element Analysis structural models are developed for generation of large datasets of high-fidelity aerodynamic loading, displacement and stress for the NREL 5 MW wind turbine blade. Based on the datasets, analysis of structural responses of the turbine blade and their possible shifts with respect to a change in the operating conditions, e.g. rotor plane tilting and yawing, are performed. A Bayesian Network model is trained to produce ranges of values and likelihoods of occurrence of abnormal structural response or the most likely damage in the blade from inputs of operating conditions. The numerical analysis showed that yawing and tilting of rotor plane has very strong influence on the performance of a wind turbine. The out-of-plane bending is the dominant mode of the rotor blade while twisting and other complex modes of displacement could be more significant during some operating conditions such as at high tilt and yaw angles. The Bayesian Network is capable of computing probability mass functions of mechanical stress or displacement of the blade given a rotor rotation speed and a type of damage or vice versa, identifying a type of damage given a measured stress distribution level.

Applied Machine Learning Predictive Analytics to SQL Injection Attack Detection and Prevention

Abstract: The back-end database is pivotal to the storage of the massive size of big data Internet exchanges stemming from cloud-hosted web applications to Internet of Things (IoT) smart devices. Structured Query Language (SQL) Injection Attack (SQLIA) remains an intruder’s exploit of choice on vulnerable web applications to pilfer confidential data from the database with potentially damaging consequences. The existing solutions of mostly signature approaches were all before the recent challenges of big data mining and at such lacks the functionality and ability to cope with new signatures concealed in web requests. An alternative Machine Learning (ML) predictive analytics provides a functional and scalable mining to big data in detection and prevention of SQLIA. Unfortunately, lack of availability of readymade robust corpus or data set with patterns and historical data items to train a classifier are issues well known in SQLIA research. In this project, we explore the generation of data set containing extraction from known attack patterns including SQL tokens and symbols present at injection points. Also, as a test case, we build a web application that expects dictionary word list as vector variables to demonstrate massive quantities of learning data. The data set is pre-processed, labelled and feature hashing for supervised learning. The trained classifier to be deployed as a web service that is consumed in a custom dot NET application implementing a web proxy Application Programming Interface (API) to intercept and accurately predict SQLIA in web requests thereby preventing malicious web requests from reaching the protected back-end database. This project demonstrates a full proof of concept implementation of an ML predictive analytics and deployment of resultant web service that accurately predicts and prevents SQLIA with empirical evaluations presented in Confusion Matrix (CM) and Receiver Operating Curve (ROC).

Short-term wind power prediction using a novel model based on butterfly optimization algorithm-variational mode decomposition-long short-term memory

The precise forecasting of wind power output is crucial for the integration of large-scale renewable energy generation into the power grid. It plays a vital role in ensuring safe and stable operation of power grid, diminishing fuel consumption, and minimizing environmental impacts. This research presents a machine learning prediction model that integrates intelligent optimization algorithms and data decomposition techniques for short-term wind power output forecasting. The initial phase of this research develops a prediction model utilizing variational mode decomposition and long short-term memory networks. To mitigate the randomness and uncertainty of wind energy, and improve prediction accuracy, the butterfly optimization algorithm is introduced to optimize the parameters of variational mode decomposition and long short-term memory networks. Several in-depth case studies are carried out on the actual wind power generation dataset in mainland China to confirm the feasibility and effectiveness of the proposed hybrid model. Experimental results demonstrate that the proposed model outperforms the compared model in terms of prediction accuracy across different seasons, showing good practicality and generalizability.