Input Dataset Research Articles

A shrinkage-free approach for fusing GRACE-based total water storage changes with models using wavelet multiresolution analysis

The observations from Gravity Recovery and Climate Experiment (GRACE) satellites provide total water storage changes (TWSC) at ∼300 km or larger spatial scales. To obtain information at a higher spatial resolution, suitable for management purposes, previous studies combined GRACE TWSC maps with information from hydrological models via assimilation techniques (e.g. CLSM-DA), or with external observations via scaling and machine learning approaches. Both techniques may introduce shrinkage errors, a reduction in amplitude of long spatial wavelength signals, and distorting water budgets at basin scale best observed by GRACE. Here, we present a fusion approach based on wavelet multiresolution analysis to combine long-wavelength GRACE TWSC observations with short-wavelength measurements of TWSC from the GLDAS Noah model in the contiguous United States (CONUS). We decompose TWSC maps into building blocks at different spatial wavelengths, examine their statistical characteristics, and combine complementary components at wavelet coefficient levels. The fused TWSC is then obtained by inverse wavelet transform. A spectral analysis indicates that the fused TWSC comprises frequency content that balances spectral characteristics of both input datasets. The fused TWSC dataset possesses enhanced details in the spatial domain, while it accurately quantifies the water budget and its long-term spatial trend at basin scale during 2003–2015, showing a good agreement with GRACE estimates. Independent validation against elastic hydrologic loading deformation measured at ∼2000 GNSS stations across CONUS shows similar overall performance for GRACE and fused datasets, while it outperforms GRACE in modeling groundwater storage changes in CONUS when compared to CLSM-DA.

Fake news detection using knowledge graph and graph convolutional network

People will increasingly get expedited and diverse means of accessing news as societies progress. Furthermore, there is a noticeable increase in the prevalence of incorrect and misleading information. Our research is motivated by the significant concerns regarding the detrimental impacts of disinformation on the general public, political stability, and trust in the media. The scarcity of Vietnamese-language datasets can be attributed to the predominant focus of false news detection studies on datasets only in English. Detection investigations of fake news have predominantly relied on supervised machine learning algorithms, which possess notable limitations when confronted with unclassified news articles that are either authentic or untrue. The utilization of Knowledge Graphs (KG) and Graph Convolutional Networks (GCN) holds promise in addressing the constraints of supervised machine learning algorithms. To address these problems, we propose an approach that integrates KG)into the procedure for detecting fake news. We utilize the Vietnamese Fake News Detection dataset (VFND-vietnamese-fake-news), comprising authentic and deceptive news articles from reputable Vietnamese newspapers such as vnexpress, tuoitre, and have been collected from 2018 to 2023. News articles are only labeled as real or fake after experiencing independent verification. The Glove embedding (Global Vectors for Word Representation) is employed to establish a knowledge network for the given dataset. This knowledge graph’s construction is accomplished using the Word Mover’s Distance (WMD) algorithm in conjunction with the K-nearest neighbor approach; GCN approach and the input KG train models to discern between real and fake news. With labeling half of the input dataset, the experimental findings indicate a notable level of accuracy, reaching up to 85%. Our research holds significant importance in identifying fake news, particularly within the context of the Vietnamese language.

Bio-Inspired Artificial Intelligence with Natural Language Processing Based on Deceptive Content Detection in Social Networking.

In recent research, fake news detection in social networking using Machine Learning (ML) and Deep Learning (DL) models has gained immense attention. The current research article presents the Bio-inspired Artificial Intelligence with Natural Language Processing Deceptive Content Detection (BAINLP-DCD) technique for social networking. The goal of the proposed BAINLP-DCD technique is to detect the presence of deceptive or fake content on social media. In order to accomplish this, the BAINLP-DCD algorithm applies data preprocessing to transform the input dataset into a meaningful format. For deceptive content detection, the BAINLP-DCD technique uses a Multi-Head Self-attention Bi-directional Long Short-Term Memory (MHS-BiLSTM) model. Finally, the African Vulture Optimization Algorithm (AVOA) is applied for the selection of optimum hyperparameters of the MHS-BiLSTM model. The proposed BAINLP-DCD algorithm was validated through simulation using two benchmark fake news datasets. The experimental outcomes portrayed the enhanced performance of the BAINLP-DCD technique, with maximum accuracy values of 92.19% and 92.56% on the BuzzFeed and PolitiFact datasets, respectively.

Machine Learning-Based Cost-Effective Smart Home Data Analysis and Forecasting for Energy Saving

This study aims to enable cost-effective Internet of Things (IoT) system design by removing redundant IoT sensors through the correlation analysis of sensing data collected in a smart home environment. This study also presents a data analysis and prediction technology that enables meaningful inference through correlation analysis of data from different heterogeneous IoT sensors installed inside a smart home for energy efficiency. An intelligent service model that can be implemented based on a machine learning algorithm in a smart home environment is proposed. Herein, seven types of sensor data are collected and classified into sets of input data (six environmental data) and target data (power data of HVAC). By using the six new input data, the power data can be predicted by the artificial intelligence model. The model performance was measured using RMSE, and the gradient-boosting regressor (gb) model performed the best, with an RMSE of 22.29. Also, the importance of sensor data is extracted through correlation analysis, and sensors with low importance are removed according to the importance of sensor values. This process can reduce costs by 13%, thereby providing a design guide for a cost-effective IoT system.

Assessing the phosphorus cycle in European agricultural soils: Looking beyond current national phosphorus budgets

Phosphorus (P) is an essential nutrient for all crops, yet its excess negatively affects public health, the environment, and the economy. At the same time, rock P is a critical raw material due to its importance for food production, the finite geological deposits, and its unequal regional distribution. As a consequence, nutrient management is addressed by numerous environmental policies. Process-based biogeochemical models are valuable instruments to monitor the P cycle and predict the effect of agricultural management policies. In this study, we upscale the calibrated DayCent model at European level using data-derived soil properties, advanced input data sets, and representative management practices. Our results depicted a P budget with an average P surplus (0.11 kg P ha−1 year−1), a total soil P (2240.0 kg P ha−1), and available P content (77.4 kg P ha−1) consistent with literature and national statistics. Through agricultural management scenarios, we revealed a range of potential changes in the P budget by 2030 and 2050, influenced by the interlink of P with biogeochemical carbon and nitrogen cycles. Thus, we developed a powerful assessment tool capable of i) identifying areas with P surplus or deficit at high spatial resolution of 1 km2, (ii) pinpointing areas where a change in agricultural management would be most urgent to reach policy goals in terms of environmental pollution, food security and resource efficiency of a critical raw material, and iii) assessing the response of the P cycle to modifications in agricultural management.

Improving the credibility of the extracted position from a vast collection of job offers with machine learning ensemble methods

The main aim of this paper is to evaluate crawlers collecting the job offers from websites. In particular the research is focused on checking the effectiveness of ensemble machine learning methods for the validity of extracted position from the job ads. Moreover, in order to significantly reduce the training time of the algorithms (Random Forests and XGBoost), granularity methods were also tested to significantly reduce the input training dataset. Both methods achieved satisfactory results in accuracy and F1 measures, which exceeded 96%. In addition, granulation reduced the input dataset by more than 99%, and the results obtained were only slightly worse (accuracy between 1% and 5%, F1 between 3% and 8%). Thus, it can be concluded that the considered methods can be used in the evaluation of job web crawlers.

Morphometrics and machine learning discrimination of the middle Eocene radiolarian species Podocyrtis chalara, Podocyrtis goetheana and their morphological intermediates

We present various approaches to distinguish the middle Eocene species Podocyrtis chalara and Podocyrtis goetheana, which are end members of a trajectory of phenotypic change, and their intermediate morphogroups. We constructed a set of thirteen traditional morphological variables to classify the entire morphological variability encompassed by the two morphospecies and their intermediates Podocyrtis sp. cf. P. chalara and Podocyrtis sp. cf. P. goetheana. We used two methods of classification, namely Linear Discriminant Analysis (LDA) and machine learning using artificial neural networks. LDA performed on the morphometric data reveals a good discrimination for P. chalara, P. goetheana and Podocyrtis sp. cf. P. goetheana, but not for Podocyrtis sp. cf. P. chalara. We used three approaches of machine learning based on different neural networks: a Convolutional Neural Network (CNN) and two Spiking Neural Networks (SNNs). Each of these neural networks was trained based on classified images of the two morphospecies and their morphological intermediates, thus constituting a different set of input data than the morphometric dataset for LDA. The neural network approaches identified the same three morphospecies recognized by LDA from a dataset of traditional measurements, i.e. P. chalara, P. goetheana and Podocyrtis sp. cf. P. goetheana, with up to 92% accuracy. Our results highlight the great potential and promising perspectives of machine learning and neural networks in the application of image-based object recognition for morphological classification, which may also contribute to more objective taxonomic decisions.

Real-Time Stress Field Prediction of Umbilical Based on PyEf-U-Net Convolutional Neural Network

Stress field analysis is an essential part of umbilical component layout design. The stress field analysis of an umbilical, via numerical simulation, has commonly been applied in practical engineering. The high economic and time cost associated with numerical simulation and analysis of the stress field in an umbilical has been replaced by data-driven, deep-learning-based, real-time computational methods. In this study, a novel Pyramidal Efficient U-Net (PyEf-U-Net) network is proposed to predict the stress field distribution of the umbilical. The input dataset is obtained via the Differential Evolution-Generalized Lagrange Multiplier (DE-GLM) method, which is entered into the network for training, with a detailed discussion of the effects of hyperparameters such as optimizer, learning rate, and loss function on the performance of the network. The experimental research demonstrates that the proposed PyEf-U-Net can accurately predict the stress field of the umbilical in real time with a prediction accuracy of 94.2%, which is superior to other deep learning networks. The proposed method can provide an effective way for rapid mechanical analysis and design of the umbilical in practical engineering, while the method can be extended to the mechanical analysis and design of other similar marine engineering equipment structures.

Deep Learning-Based Wildfire Image Detection and Classification Systems for Controlling Biomass

Forests are essential natural resources that directly impact the ecosystem. However, the rising frequency of forest fires due to natural and artificial climate change has become a critical issue. A revolutionary municipal application proposes deploying an artificial intelligence-based forest fire warning system to prevent major disasters. This work aims to present an overview of vision-based methods for detecting and categorizing forest fires. The study employs a forest fire detection dataset to address the classification difficulty of discriminating between photos with and without fire. This method is based on convolutional neural network transfer learning with Inception-v3. Thus, automatic identification of current forest fires (including burning biomass) is a critical field of research for reducing negative repercussions. Early fire detection can also assist decision-makers in developing mitigation and extinguishment strategies. Radial basis function Networks (RBFNs) with rapid and accurate image super resolution (RAISR) is a deep learning framework trained on an input dataset to detect active fires and burning biomass. The proposed RBFN-RAISR model’s performance in recognizing fires and nonfires was compared to earlier CNN models using several performance criteria. The water wave optimization technique is used for image feature selection, noise and blurring reduction, image improvement and restoration, and image enhancement and restoration. When classifying fire and no-fire photos, the proposed RBFN-RAISR fire detection approach achieves 97.55% accuracy, 93.33% F-Score, 96.44% recall, 94.19% precision, and an error rate of 24.89. Given the one-of-a-kind forest fire detection dataset, the suggested method achieves promising results for the forest fire categorization problem.

A Novel Clustering Method Based on Adjacent Grids Searching.

Clustering is used to analyze the intrinsic structure of a dataset based on the similarity of datapoints. Its widespread use, from image segmentation to object recognition and information retrieval, requires great robustness in the clustering process. In this paper, a novel clustering method based on adjacent grid searching (CAGS) is proposed. The CAGS consists of two steps: a strategy based on adaptive grid-space construction and a clustering strategy based on adjacent grid searching. In the first step, a multidimensional grid space is constructed to provide a quantization structure of the input dataset. The noise and cluster halo are automatically distinguished according to grid density. Moreover, the adaptive grid generating process solves the common problem of grid clustering, in which the number of cells increases sharply with the dimension. In the second step, a two-stage traversal process is conducted to accomplish the cluster recognition. The cluster cores with arbitrary shapes can be found by concealing the halo points. As a result, the number of clusters will be easily identified by CAGS. Therefore, CAGS has the potential to be widely used for clustering datasets with different characteristics. We test the clustering performance of CAGS through six different types of datasets: dataset with noise, large-scale dataset, high-dimensional dataset, dataset with arbitrary shapes, dataset with large differences in density between classes, and dataset with high overlap between classes. Experimental results show that CAGS, which performed best on 10 out of 11 tests, outperforms the state-of-the-art clustering methods in all the above datasets.

Construction of a feedback control system based on CFD simulations for the 64-element Canadian SCWR

The knowledge of the dynamic behaviors of the supercritical water in the fuel bundle of the nuclear reactor is necessary for the design of the control system. For this purpose, the transient full-scale computational fluid dynamics (CFD) simulations are used to obtain the dynamic input and output data sets of the 64-element Canadian supercritical water-cooled reactor (SCWR) in this study. Subsequently, the linear dynamic models of the reactor are obtained by the system identification technique and validated by the comparison with the results from CFD simulations. Based on the linear dynamic models, a feedback control system consists of three PI/PID controllers is constructed. The objective is to regulate the system back to the design point when the reactor is subjected to disturbances. Therefore, the performance of the designed feedback control system is also evaluated in this work. It shows that the designed control system can timely minimize the deviations and return the reactor back to the desired operating condition.

Machine learning‐based regional flood frequency analysis of Indian watersheds

AbstractEstimates of the flood quantile for ungauged watersheds are crucial for water resources management but challenging due to the nonlinear complex hydrological system. For ungauged watersheds, estimating flood quantiles relies on various interdependent physiometeorological variables, many of which are not adequately considered in regional flood frequency analysis (RFFA). In this study, we utilized the random forest (RF) and support vector regression (SVR) algorithms, which can learn the nonlinear relationship between the physiometeorological variables and flood quantiles for RFFA. Thirteen physiometeorological variables that were not collectively employed before were used to estimate the 10‐year, 50‐year, and 100‐year return period flood quantiles (Q10, Q50, and Q100), respectively, for 39 watersheds spread across India. The RF and SVR models were trained on 29 (75%) watersheds to estimate individual flood quantiles and were subsequently tested on the remaining ten (25%) ungauged watersheds. The R2 achieved by RF is 0.862, 0.813, and 0.845, and SVR is 0.807, 0.793, and 0.789 for Q10, Q50, and Q100, respectively. Overall, the results indicate that RF can effectively learn the nonlinear relationships, while SVR with a linear kernel requires further improvement to estimate reliable flood quantiles. The study demonstrates that machine learning algorithms, with appropriate physiometeorological input datasets, can be used to estimate flood quantiles even in the sparse data region.

An Adaptive Decision Tree Regression Modeling for the Output Power of Large-Scale Solar (LSS) Farm Forecasting

The installation of large-scale solar (LSS) photovoltaic (PV) power plants continues to rise globally as well as in Malaysia. The data provided by LSS PV consist of five weather stations with seven parameters, a 22-unit inverter, and 1-unit PQM Meter Grid as a big dataset. These big data are rapidly changing every minute, they lack data quality when missing data, and need to be analyzed for a longer duration to leverage their benefits to prevent misleading information. This paper proposed the forecasting power LSS PV using decision tree regression from three types of input data. Case 1 used all 35 parameters from five weather stations. For Case 2, only seven parameters were used by calculating the mean of five weather stations. While Case 3 was chosen from an index correlation of more than 0.8. The analysis of the historical data was carried out from June 2019 until December 2020. Moreover, the mean absolute error (MAE) was also calculated. A reliability test using the Pearson correlation coefficient (r) and coefficient of determination (R2) was done upon comparing with actual historical data. As a result, Case 2 was proposed to be the best input dataset for the forecasting algorithm.

Geometry Representation Effectiveness in Improving Airfoil Aerodynamic Coefficient Prediction with Convolutional Neural Network

Many applications use symmetric or asymmetric airfoils, such as aircraft design, wind turbines, and heat transfer. Each airfoil has different aerodynamic coefficients. Obtaining the aerodynamic coefficients is a must to optimize the airfoil design. Engineers use various methods to get the airfoil aerodynamic coefficients. A prediction method is an approximation approach that effectively reduces time and cost. This article uses convolutional neural networks (CNN) to get approximation values of those coefficients. In CNN, we collect 8920 aerodynamic coefficients for 223 NACA 4 as labels in datasets by using XFOIL at  and  with varying angles of attacks starting  to  with increment of . The simulation results are compared to the experiment using E387 airfoil for validation. Then, airfoil geometries as part of input datasets were transformed into Grayscale and RGB images using the signed distance function (SDF) and mesh algorithm. Each airfoil representation was trained using an 80% dataset and tested using a 20% dataset with Adam as an optimizer to generate each prediction model using modified LeNet-5. We use three different layer depths in modified LeNet-5 to obtain the optimal layer number. There is no remarkable improvement when varying the depth layers, so four layers are used instead. Simulation results show that using an SDF with Fast Marching Method on CNN predicts the most effective for the airfoil’s lift, drag, and pitch moment coefficient with varying angles of attack simultaneously. One can extend the method by using SDF to recognize different flow conditions.

Mitigating underreported error in food frequency questionnaire data using a supervised machine learning method and error adjustment algorithm

BackgroundFood frequency questionnaires (FFQs) are one of the most useful tools for studying and understanding diet-disease relationships. However, because FFQs are self-reported data, they are susceptible to response bias, social desirability bias, and misclassification. Currently, several methods have been created to combat these issues by modelling the measurement error in diet-disease relationships.MethodIn this paper, a novel machine learning method is proposed to adjust for measurement error found in misreported data by using a random forest (RF) classifier to label the responses in the FFQ based on the input dataset and creating an algorithm that adjusts the measurement error. We demonstrate this method by addressing underreporting in selected FFQ responses.ResultAccording to the results, we have high model accuracies ranging from 78% to 92% in participant collected data and 88% in simulated data.ConclusionThis shows that our proposed method of using a RF classifier and an error adjustment algorithm is efficient to correct most of the underreported entries in the FFQ dataset and could be used independent of diet-disease models. This could help nutrition researchers and other experts to use dietary data estimated by FFQs with less measurement error and create models from the data with minimal noise.

Prediction of biocrude oil yields from hydrothermal liquefaction using a gradient tree boosting machine approach with principal component analysis

Biomass can be converted to bio-fuels and bio-chemicals via many thermo-chemical conversion platforms such as hydrothermal liquefaction (HTL) which occurs at high temperature and pressure conditions. HTL can be used for converting wet or high moisture content biomass to biocrude oils effectively. There are wide variety of relevant parameters affecting the system of biomass HTL, for example, operating conditions, biomass characteristics, solvent properties, or catalyst identities used. This brings about the difficulty in making highly accurate prediction of biocrude oil yields from biomass HTL. In this work, a gradient tree boosting machine learning (GTB-ML) model with principal component analysis (PCA) was applied to predict the yields of biocrude oils from biomass HTL, using 15 input variables from biomass characteristics, operating conditions, and solvent properties. PCA is a statistical method to find correlations in multivariate data, and it can be used to specify highly influential variables in input datasets for ML models. PCA results confirmed that input variables from biomass characteristics were as important as those from HTL operating conditions. The GTB-ML developed from principal components 1 to 8, representing about 90% of the whole original dataset can be used to predict the yields of bio-crude oils from biomass HTL at acceptable accuracy with correlation coefficient, R2=0.8 and root mean square error = 0.005.

A PPA Study of Reinforced Placement Parameter Autotuning: Pseudo-3D vs. True-3D Placers

D Place and Route (P&R) flows either involve true-3D placement algorithms or use commercial 2D tools to transform a 2D design into a 3D design. Irrespective of the nature of the placers, several placement parameters in these tools affect the quality of the final 3D designs. Different parameter settings work well with different circuits, and it is impossible to manually tune them for a particular circuit. Automated approaches involving reinforcement learning have been shown to adapt and learn the parameter settings and create trained models. However, their effectiveness depends on the input dataset quality. Using a set of 10 netlists and 10–21 handpicked placement parameters in P&R flows involving pseudo-3D or true-3D placement, the dataset quality is analyzed. The datasets are the design metrics obtained through different P&R stages, such as placement optimization, clock tree synthesis, or 3D partitioning and global routing. The training runtime and the quality of the final design metrics are compared. On a pseudo-3D flow, the training takes around 126–290 hours, whereas, on a true-3D placer-based flow, it takes around 305–410 hours. It is observed that the datasets obtained from different stages lead to drastically different final design results. With the RL-based training processes, the quality of results in 3D designs improves by up to 23.7% compared to their corresponding untrained P&R flows.

Dynamic Obstacle Avoidance Technique for Mobile Robot Navigation Using Deep Reinforcement Learning

In the realm of mobile robotics, navigating around obstacles is a fundamental task, particularly in constantly changing situations. Although deep reinforcement learning (DRL) techniques exist that utilize the positional information of robot’s, environmental states, and input dataset for neural networks. Although, the positional information alone does not provide sufficient insights into the motion trends of obstacles. To solve this issue, this paper presents a dynamic obstacle mobility pattern approach for mobile robots (MRs) that rely on DRL. This method employs the positional details of dynamic obstacles dependent upon time for establishing a movement trend vector. This vector, in conjunction with another mobility state attribute, forms the MR mobility guidance matrix, that essentially conveys the pattern variation of dynamic obstacles trend over a specified interval. Using this matrix, the robot can choose its avoidance action. Also, this methodology uses the DRL-based dynamic policy algorithm for the testing and validation of the proposed technique through Python programming. The experimental outcomes demonstrate that this technique substantially improves the safety of avoiding dynamic obstacles

Soil Texture Prediction Using Machine Learning Approach for Sustainable Soil Health Management

Soil in the earth acts as a foothold for all crops. Soil texture is the most important soil health indicator being used for the selection of crops, mechanical manipulation, irrigation management, and fertilizer management. The texture of the soil influences the storage and flow of air and water within the soil, as well as root development, the accessibility of plant nutrients, and the activities of different microorganisms. These factors collectively impact the soil's fertility, quality, and soil health. A conventional method of soil texture analysis is cumbersome, time-consuming, and labor-intensive. Machine learning (ML) is a newly emerging technique being used to assess the soil's physical, chemical, and biological properties quickly in real-time. This is an eco-friendly approach since it does not involve any hazardous chemicals. Machine learning can learn complex features and predict nonlinear properties. Convolutional Neural Networks (CNN) employs convolutional layers to automatically learn features from the input data and is widely used in image classification, object detection, and image generation tasks in a short time. Soil texture images are given as input dataset after the completion of image subsetting, data preprocessing, and Image augmentation. This gives a CNN-based soil texture predictive model with a reliable accuracy of 87.50% at a lower cost.

Morphometric analysis of ice scour lakes in Iceland: A proxy for ice sheet dynamics

AbstractGlacial erosion rates on the Iceland landscape throughout the Pleistocene are not well quantified. Ice scour lakes provide an opportunity to investigate glacial erosive activity and relative intensity because they commonly form in areas beneath ice sheets due to intense quarrying. This study evaluates ice scour lake morphology and density as a potential proxy for paleo‐ice flow direction and basal thermal regime for parts of the Iceland Ice Sheet. Using GIS analysis, properties of ice scour lakes are examined in regions that experienced different rates of ice flow during and following the Last Glacial Maximum (LGM), as interpreted from streamlined subglacial landforms. The primary input datasets were topographic, hydrologic and bedrock data. Lake distribution and morphology were quantified for all natural lakes of Iceland (n = 30 169), with close examination of three 400 km2 sub‐regions in Vestfirðir (n = 904), Húnaflói (n = 69) and Bakkaheiði (n = 232). Lake distribution parameters include density, packing and centroid elevation. PolyMorph‐2D was used to quantify lake morphology including orientation, area and elongation ratio. Ice scour lake density, packing and elevation were all greatest on Vestfirðir. The high density and packing of lakes across Vestfirðir suggests unique ice dynamics relative to the other two sub‐regions, and multidirectional flow of lake axes supports the presence of an independent ice cap covering Vestfirðir. Differential intensity of glacial erosion interpreted from regions of high and low lake density on Bakkaheiði supports a proposed ice divide in Northeast Iceland. The orientation of lakes align with the flow directions of proposed LGM ice streams in Northern and Northeast Iceland. Ice scour lakes were most elongate on average on Bakkaheiði, which supports fast ice flow. Improving understanding of former ice sheet basal thermal regime and paleo‐ice flow velocity serves to inform modern controls on ice sheet stability and ice sheet basal processes.