Search

Artificial Intelligence is quickly emerging as a technological solution for the agriculture industry to surmount its classical challenges. Artificial Intelligence is facilitating farmers to refine their products and alleviate unfavourable impacts due to the environment. The central concern of this paper is predictive analytics to develop a machine learning model to identify and predict crop yield based on multiple environmental factors. In this paper, a hybrid learner 'RaNN' is proposed that combines the feature sampling and majority voting technique of Random Forest in-combination with the multilayer Feedforward Neural Network to predict the crop yield. Research has also ascertained the essential features responsible for accurate yield prediction. The proposed model works for rice yield prediction, one of the chief grains of India. The region chosen for the work is Punjab, which is among the largest producer states of India for rice. The dataset consists of 15 attributes comprising the weather and agriculture data collected from the Indian Meteorological Department Pune, and Punjab Environment Information System (ENVIS) Center, Government of India. The study has also made a comparative assessment of 'RaNN' with machine learning methods like Multiple Linear Regression, Random Forest, Decision Tree, Boosting Regression, Support Vector Machine Regression, Ensemble Learner, and Artificial Neural Network. Our model RaNN has listed a better prediction accuracy with minimal error among the other techniques providing a 98% correlation between the actual and the predicted yield. Abbreviations: AI – Artificial Intelligence; ANN – Artificial Neural Network; BR – Boosting Regression; Chem Fert – chemical fertilisers; DT – Decision Tree; EL – Ensemble Learner; ENVIS – Punjab Environment Information System; GBM – Stochastic Gradient Boosting Method; GPS – Global Positioning System; HMAX – highest maximum temperature in degrees C; IMD – Indian Meteorological Department; L1 – Lasso regression; L2 – Ridge regression; LMIN – lowest minimum temperature; ML – Machine Learning; MAE – Mean Absolute Error; MEVP – mean evaporation in mm; MLR – Multiple Linear Regression; MMAX – mean maximum temperature in degrees C; MMIN – mean minimum temperature in degrees C; MSSH – Mean sunshine duration in hours; MWS – mean wind speed in km/h; P1 – number of days with precipitation (0.1–0.2 mm); P2 – number of days with precipitation (greater than or equal to 0.3 mm); RaNN – Hybrid RF-ANN model; RMSE – Root Mean Squared Error; $${R^2}$$ R 2 – Coefficient of determination; RD – number of rainy days; RF – Random Forest; SVM Reg – Support Vector Machine Regression; TMRF – total rainfall per month in mm [ABSTRACT FROM AUTHOR]

In this paper, the spoke structure of a non-pneumatic tire is optimized. The finite element model of the spoke structure is created and structural dynamic analysis is made using ABAQUS. The optimization goal is to simultaneously reduce three characteristics of the spoke structure, the mass, vertical displacement and acceleration. Simulations are performed according to the center composite design table, and the objective function is estimated using response surface analysis (RSA), random forest regression (RFR), gradient boosting regression (GBR) and artificial neural network (ANN). The genetic algorithm is used for minimization, and the optimization results of each objective function are analyzed through verification simulation. As a result of the optimization, the error between the predicted value and the actual value is 18.6% for RSA 3.1% for RFR, 1.7% for GBR, and 15.0% for the ANN. The ANN proposes the optimum design variables which derive output values smaller than the current minimum value because the ANN properly represents the nonlinear characteristics of the spoke model generated in this study. Highlights Simulation based optimization of the spoke structure of a non-pneumatic tire Various regression models are used and their accuracy is compared A two-step genetic algorithm is used for optimization [ABSTRACT FROM AUTHOR]

The development of fragility curves is an important step in seismic risk assessment within the scope of performance‐based earthquake engineering. The goal of this work is to use machine learning methods (regression‐based tools) to forecast the large‐span precast parking structural responses and fragility curves. This study proposes five predicting models based on machine learning to evaluate the seismic performance of the large‐span precast parking structures, including: neural networks, genetic algorithm‐based neural networks, support vector machine, decision tree and random forest. A database that includes 453 numerical synthetic results was used to train and test the machine learning models. The seismic performance of large‐span precast parking structures were predicted using the constructed machine learning models. Finally, the sensitivity analysis of input parameters was conducted. From this paper we can conclude that: (1) the genetic optimization‐based neural networks' predicting model has the most accurate predictive ability for seismic fragility estimation and (2) the structural responses and the fragility curves of parking structures are related to the differences of the stiffness of the connectors and the number of floors, of which the stiffness of the connectors should be given special attention. [ABSTRACT FROM AUTHOR]

The paper explores the possibilities of using data classification methods when forecasting time series of the geomagnetic Kp-index by machine learning methods. To classify categories of the Kp-index based on the degree of disturbance, linear and logistic regression, random forest, gradient boosting on top of decision trees, and artificial neural networks of various architectures are used. The results of these methods are compared with a trivial inertial forecast (the statistical indicators of which for problems of this type are always high) at horizons from 3 h to 1 day in 3-h increments. The problem of choosing a cross-validation scheme for selecting the model hyperparameters, ways to overcome the imbalance of categories, the relative importance of input features, as well as the dependence of the results on the test sample (beginning of the 25th solar activity cycle) on inclusion in the training sample of data from the 23rd and 24th cycles or only the 24th cycles are studied. Based on the results, conclusions are drawn about the preferred methods for classifying values of the Kp-index based on the level of geomagnetic disturbance. Ways for further research and possible improvement of the classification quality are outlined, including for determining the characteristic hidden states of Earth's magnetosphere as a dynamic system in order to improve the quality of forecasting geomagnetic indices. [ABSTRACT FROM AUTHOR]

This paper aims to build a machine learning and a neural network model to detect the probability of money laundering in banks. The paper's data came from a simulation of actual transactions flagged for money laundering in Middle Eastern banks. The main findings highlight that criminal networks mainly use the integration stage to integrate money into the financial system. Fraudsters prefer to launder funds in the early hours, morning followed by the business day's afternoon time intervals. Additionally, the Naïve Bayes and Random Forest classifiers were identified as the two best-performing models to predict bank money laundering transactions. [ABSTRACT FROM AUTHOR]

This paper explores the different insider employee-led cyber frauds (IECF) based on the recent large-scale fraud events of prominent Indian banking institutions. Examining the different types of fraud and appropriate control measures will protect the banking industry from fraudsters. In this study, we identify and classify Cyber Fraud (CF), map the severity of the fraud on a scale of priority, test the mitigation effectiveness, and propose optimal mitigation measures. The identification and classification of CF losses were based on a literature review and focus group discussions with risk and vigilance officers and cyber cell experts. The CF was analyzed using secondary data. We predicted and prioritized CF based on machine learning-derived Random Forest (RF). An efficient fraud mitigation model was developed based on an offender-victim-centric approach. Mitigation is advised both before and after fraud occurs. Through the findings of this research, banks and fraud investigators can prevent CF by detecting it quickly and controlling it on time. This study proposes a structured, sustainable CF mitigation plan that protects banks, employees, regulators, customers, and the economy, thus saving time, resources, and money. Further, these mitigation measures will improve the reputation of the Indian banking industry and ensure its survival. [ABSTRACT FROM AUTHOR]

Malaria continues to be a major barrier to socioeconomic development in Africa, where its death rate is over 90%. The predictive power of many machine learning models—such as multi-linear regression (MLR), artificial neural networks (ANN), adaptive neuro-fuzzy inference systems (ANFISs) and Random Forest classifier—is investigated in this study using data from 2207 patients. The dataset was reduced from the initial dataset of thirty-two criteria samples to fifteen. Assessment measures such as the root mean square error (RMSE), mean square error (MSE), coefficient of determination (R2), and adjusted correlation coefficient R were used. ANFIS, Random Forest, MLR, and ANN are among the models. After training, ANN outperforms ANFIS (97%), MLR (92%), and Random Forest (68%) with the greatest R (99%) and R2 (99%), respectively. The testing stage confirms the superiority of ANN. The paper also presents a statistical forecasting sheet with few errors and excellent accuracy for MLR models. When the models are assessed with Random Forest, the latter shows the least results, thus broadening the modeling techniques and offering significant insights into the prediction of malaria and healthcare decision making. The outcomes of using machine learning models for precise and efficient illness prediction add to an expanding body of knowledge, assisting healthcare systems in making better decisions and allocating resources more effectively. [ABSTRACT FROM AUTHOR]

This paper analyzes the efficiency of various machine learning models (artificial neural networks, random forest, decision tree, AdaBoost and XGBoost) against the evolution of VBA-based (Visual Basic for Applications) malware over a large period of time (1995–2021). The file set used in our research is comprehensive—approximately 1.9 million files (out of which 944,595 are malicious and the rest are benign)—which allowed to gain insights on the resilience of various machine learning models against the diversity and the evolution of file features that reflect obfuscation techniques in VBA-based malware. In studying detection of VBA-based malware, we focus on characteristics of both the classifiers—proactivity (short-term detection efficiency against future malware), endurance (long-term detection robustness)—and of the detection-wise relevant file features—feature perishability (dynamics of feature relevance). We also describe in some detail—as a prerequisite of the study—various obfuscation techniques used by the malware under investigation during the last decade. [ABSTRACT FROM AUTHOR]

Intensity modulation/direct detection (IM/DD) optical interconnect systems are susceptible to waveform distortion caused by complex nonlinearities. To reduce the effect of waveform distortion on signal transmission accuracy, in this paper, a random forest (RF) equalization algorithm based on decision tree (DT) integration is proposed for cost-sensitive IM/DD systems, which has the characteristics of nonlinear classification and regression. The performance of the RF equalizer is verified by simulation experiments on a 120 Gbps 8-level pulse amplitude modulation (PAM8) transmission system using an electro-absorption-modulated laser (EML). Simulation studies show that whether it is a back-to-back (B2B) transmission system or a 10 km optical fiber transmission system, the RF equalization scheme can achieve much better bit error rate (BER) performance than the traditional equalization scheme. In addition, in both transmission cases, the RF equalization scheme enables efficient classification of signals and is much less simpler than the artificial neural network (ANN) equalization scheme. [ABSTRACT FROM AUTHOR]

In recent years, deep neural networks have evolved rapidly in engineering technology, with models becoming larger and deeper. However, for most companies, developing large models is extremely costly and highly risky. Researchers usually focus on the performance of the model, neglecting its cost and accessibility. In fact, most regular business scenarios do not require high-level AI. A simple and inexpensive modeling method for fulfilling certain demands for practical applications of AI is needed. In this paper, a Fragmented neural network method is proposed. Inspired by the random forest algorithm, both the samples and features are randomly sampled on image data. Images are randomly split into smaller pieces. Weak neural networks are trained using these fragmented images, and many weak neural networks are then ensembled to build a strong neural network by voting. In this way, sufficient accuracy is achieved while reducing the complexity and data volume of each base learner, enabling mass production through parallel and distributed computing. By conducting experiments on the MNIST and CIFAR10 datasets, we build a model pool using FNN, CNN, DenseNet, and ResNet as the basic network structure. We find that the accuracy of the ensemble weak network is significantly higher than that of each base learner. Meanwhile, the accuracy of the ensemble network is highly dependent on the performance of each base learner. The accuracy of the ensemble network is comparable to or even exceeds that of the full model and has better robustness. Unlike other similar studies, we do not pursue SOTA models. Instead, we achieved results close to the full model with a smaller number of parameters and amount of data. [ABSTRACT FROM AUTHOR]

Natural streamflow data is required in many hydrological applications. However, many basins are located in data-scarce regions or are impacted by human construction and activities. In this paper, we explore three machine learning algorithms, namely artificial neural networks, random forest and light gradient boosting machine, to simultaneously estimate all the parameters of the coupled modèle du Génie Rural à 4 paramètres Journaliers (GR4J) and snow accounting routine called CemaNeige model. A database of 675 basins in the USA and Quebec is used to train and test ensembles. After using the estimated parameters in GR4J, the resulting naturalized streamflow series are compared with those obtained by the established drainage area ratio and spatial proximity transfer methods in 11 test basins. The results indicate that the machine learning algorithms outperform the drainage area ratio and spatial proximity transfer methods. Among machine learning algorithms, random forests obtain lower (better) continuous ranked probability scores than the other methods for 10 out of 11 test basins. [ABSTRACT FROM AUTHOR]

Driving behavior monitoring by smartphone sensors is one of the most investigated approaches to ameliorate road safety. Various methods are adopted in the literature; however, to the best of our knowledge, their robustness to the prediction of new unseen data from different drivers and road conditions is not explored. In this paper, a two-phase Machine Learning (ML) method with taking advantage of high-pass, low-pass, and wavelet filters is developed to detect driving brakes and turns. In the first phase, accelerometer and gyroscope filtered time series are fed into Random Forest and Artificial Neural Network classifiers, and the suspicious intervals are extracted by a high recall. Following that, in the next phase, statistical features calculated based on the obtained intervals are used to determine the false and true positive events. To compare the predicted and real labels of the recorded events and calculate the accuracy, a method that covers the limitations of previous sliding windows is also employed. Real-world experimental result shows that the proposed method can predict new unseen datasets with average F1-scores of 71% in brake detection and 82% in turn detection which is comparable with previous works. Moreover, by sensitivity analysis of our proposed model, it is proven that implementing high-pass and low-pass filters can affect the accuracy for turn detection up to 30%. [ABSTRACT FROM AUTHOR]

Machine-learning (ML) techniques have bloomed in recent years, especially in fluid mechanics applications. In this paper, we trained, validated and compared two types of ML-based models to augment Reynolds-averaged Navier–Stokes (RANS) simulations. The methodology was tested in a series of flows around bumps, characterized by different levels of flow separation and curvatures. Initially, the ML-based models were trained in three configurations presenting attached flow, small and moderate separation and tested in two configurations presenting incipient and large separation. The output quantity of the machine-learning model is the turbulent viscosity as done in Volpiani et al. (2022). The new models based on artificial neural networks (NN) and random forest (RF) improved the results if compared to the baseline Spalart–Allmaras model, in terms of velocity field and skin-friction profiles. We noted that NN has better extrapolation properties than RF, but the skin-friction distribution can present small oscillations when using certain input features. These oscillations can be reduced if the RF model is employed. One major advantages of RF is that raw quantities can be given as input features, avoiding normalization issues (such as division by zero) and allowing a larger number of universal inputs. At the end, we propose a mixed NN-RF model that combines the strengths of each method and, as a result, improves considerably the RANS prediction capability, even for a case with strong separation where the Boussinesq hypothesis (and therefore the eddy-viscosity assumption) lacks accuracy. • We combine neural networks and random forest to create a new data-driven RANS model. • The new machine learning-based model does not require to transport unclosed terms. • The new model improves the velocity, wall-pressure and skin-friction distributions. • It was proved to be robust even in predicting extrapolated flows. [ABSTRACT FROM AUTHOR]

For a long time, elucidating the tectonic setting of unknown rock samples has been a focal point for geologists. Traditional methodologies for this purpose have been scrutinized increasingly due to their inherent limitations. In response to these challenges, this paper applies modern machine learning techniques to analyze the geochemical data of igneous rocks and improve understanding of tectonic settings. By employing a variety of machine learning models, including Decision Trees, K-Nearest Neighbors, Support Vector Machines, Random Forests, Extreme Gradient Boosting, and Artificial Neural Networks, and training with 23 features comprising nine major elements (SiO2, TiO2, Al2O3, CaO, MgO, MnO, Na2O, K2O, and P2O5) along with 14 trace elements (La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, and Lu), the study successfully distinguished between seven different tectonic settings. Among these models, Random Forest, Extreme Gradient Boosting, and Artificial Neural Networks demonstrated superior classification accuracy and recall rates, with accuracies of 0.85, 0.87, and 0.86, respectively. This validates the effectiveness and potential of machine learning technologies in distinguishing the tectonic settings of igneous rocks through their geochemical elements. To enable geologists and researchers to more accurately understand and predict the origins of igneous rocks without the need to master machine learning knowledge, a user-friendly software, GeoTectAI, has been developed. [ABSTRACT FROM AUTHOR]

• A comprehensive, explainable accident-data-based AI artifact design is presented. • Utilize aggregate SHAP scores to classify crash data into high-level causal factors. • A systems-level taxonomy of driver behaviors linked to injury severity is discussed. • Separating explanation from models yields managerial insights. Automobile traffic accidents represent a significant threat to global public safety, resulting in numerous injuries and fatalities annually. This paper introduces a comprehensive, explainable artificial intelligence (XAI) artifact design, integrating accident data for utilization by diverse stakeholders and decision-makers. It proposes responsible, explanatory, and interpretable models with a systems-level taxonomy categorizing aspects of driver-related behaviors associated with varying injury severity levels, thereby contributing theoretically to explainable analytics. In the initial phase, we employed various advanced techniques such as data missing at random (MAR) with Bayesian dynamic conditional imputation for addressing missing records, synthetic minority oversampling technique for data imbalance issues, and categorical boosting (CatBoost) combined with SHapley Additive exPlanations (SHAP) for determining and analyzing the importance and dependence of risk factors on injury severity. Additionally, exploratory feature analysis was conducted to uncover hidden spatiotemporal elements influencing traffic accidents and injury severity levels. We developed several predictive models in the second phase, including eXtreme Gradient Boosting (XGBoost), random forest (RF), deep neural networks (DNN), and fine-tuned parameters. Using the SHAP approach, we employed model-agnostic interpretation techniques to separate explanations from models. In the final phase, we provided an analysis and summary of the system-level taxonomy across feature categories. This involved classifying crash data into high-level causal factors using aggregate SHAP scores, illustrating how each risk factor contributes to different injury severity levels. [ABSTRACT FROM AUTHOR]

Accurate prediction of critical heat flux (CHF) is important for enhancing the reliability and efficiency of systems used in the energy and power engineering sectors. Traditional physical and mathematical models are constrained by current theoretical maturity. While machine learning (ML) techniques offer alternative ways for CHF modeling, its 'black-box' nature poses risks of producing outcomes that could potentially violate physical laws. In this paper, physics-informed ML (PIML) algorithms amalgamating characteristics of both ML and physical models are investigated. Three physical models for computing CHF (Zuber, CHF Look-Up Table (LUT), and Biasi) were paired with three ML models (Neural Network, Random Forest, Convolutional Neural Network) and trained and validated using the available CHF database. The uncertainty quantification (UQ) study was also conducted to assess the influence of fluctuating inputs on output's stability. Results indicated that PIML algorithms generally had lower rRSMEs than pure data-driven ML, showing higher prediction accuracy. The results of PIML also reflect physical principles more coherently. Among PIML models, LUT-informed Neural Network and LUT-informed Random Forest algorithms balanced computational accuracy with physical fidelity. UQ study revealed that under uncertain inputs, the LUT-informed Neural Network model yielded more concentrated and stable outcomes than LUT-informed Random Forest, evidencing strong robustness. The PIML method shows great potential in CHF prediction. • Physics-informed ML's performance in predicting CHF is compared against pure ML. • Results from physics-informed ML resonate more with theoretical physics, optimizing the "black-box" feature. • LUT-informed RF stands out for CHF prediction accuracy and its alignment with physical laws. • Uncertainty quantification findings pinpoint LUT-informed RF's challenges with uncertain inputs, signaling notable output distribution bias. [ABSTRACT FROM AUTHOR]

The failure of concrete for hydraulic structures due to the excessive abrasion depth has caused great economic losses. Despite the tremendous work that has been done, the durability design of hydraulic concrete structures concerning abrasion damage is still a difficult task, especially with a long service life aiming at economic benefit and environmental sustainability. In this paper, machine learning methodologies including random forests (RFs) and artificial neural networks (ANNs), are used to establish prediction models based on concrete mixture proportions, curing age, and hydraulic conditions. Furthermore, RF models are coupled with multiple objective particle swarm optimization (MOPSO) algorithms to perform computational design optimization of concrete mixtures. The optimal solutions are generated based on 5000 randomly generated mixture proportions, among which three optimal designs of concrete mixtures are selected to guide field applications on account of concrete durability requirements and economic benefits. [ABSTRACT FROM AUTHOR]